# chatgpt_zeroshot # cola ## 10 prompts Acc: 78.50%, prompt: Review the sentence below and identify whether its grammar is 'Acceptable' or 'Unacceptable': Acc: 78.50%, prompt: Assess the grammatical structure of the given sentence and classify it as 'Acceptable' or 'Unacceptable': Acc: 77.50%, prompt: Analyze the provided sentence and classify its grammatical correctness as 'Acceptable' or 'Unacceptable': Acc: 77.50%, prompt: Check the grammar of the following sentence and indicate if it is 'Acceptable' or 'Unacceptable': Acc: 77.00%, prompt: Examine the given sentence and decide if it is grammatically sound. Answer with either 'Acceptable' or 'Unacceptable': Acc: 76.50%, prompt: Assess the following sentence and determine if it is grammatically correct. Respond with 'Acceptable' or 'Unacceptable': Acc: 76.00%, prompt: Please evaluate the grammatical structure of the provided sentence and answer with 'Acceptable' or 'Unacceptable': Acc: 76.00%, prompt: Examine the sentence and decide if its grammar is 'Acceptable' or 'Unacceptable': Acc: 75.50%, prompt: Determine if the grammar of the given sentence is 'Acceptable' or 'Unacceptable': Acc: 73.50%, prompt: Is the provided sentence grammatically correct? Respond with 'Acceptable' or 'Unacceptable': Acc: 79.50%, prompt: Functioning as a grammar evaluation tool, analyze the given sentence and decide if it is grammatically correct, responding with 'acceptable' or 'unacceptable': Acc: 79.50%, prompt: While performing grammar analysis, classify the grammar of the following sentence as 'acceptable' for correct grammar or 'unacceptable' for incorrect grammar: Acc: 79.00%, prompt: Acting as a grammar detection instrument, determine if the provided sentence is grammatically sound, answering with 'acceptable' for correct grammar or 'unacceptable' for incorrect grammar: Acc: 79.00%, prompt: As a tool for determining grammatical correctness, review the sentence and categorize its grammar as either 'acceptable' or 'unacceptable': Acc: 79.00%, prompt: Working as a grammar classifier, identify whether the given sentence has correct grammar, and respond with 'acceptable' for correct sentences or 'unacceptable' for incorrect ones: Acc: 79.00%, prompt: In the role of a syntax analyzer, examine the grammar of the provided sentence and decide if it is correct, answering with 'acceptable' for grammatically correct sentences or 'unacceptable' for incorrect ones: Acc: 78.50%, prompt: As an instrument for grammar evaluation, consider the sentence and determine if its grammar is correct, responding with 'acceptable' for correct grammar or 'unacceptable' for incorrect grammar: Acc: 77.50%, prompt: In your role as a grammar check tool, assess the following sentence and classify it as 'acceptable' if it is grammatically correct or 'unacceptable' if it is incorrect: Acc: 77.50%, prompt: In the capacity of a grammar assessment system, indicate if the structure of the provided sentence is grammatically correct, responding with 'acceptable' or 'unacceptable': Acc: 77.00%, prompt: As a grammar identification system, examine the provided sentence and respond with 'acceptable' for grammatically correct sentences or 'unacceptable' for incorrect ones: ## bertattack Original prompt: Review the sentence below and identify whether its grammar is 'Acceptable' or 'Unacceptable': Attacked prompt: b"review the citation below and do must its grammar is 'Acceptable' or 'Unacceptable':" Original acc: 78.50%, attacked acc: 43.50%, dropped acc: 35.00% Original prompt: Analyze the provided sentence and classify its grammatical correctness as 'Acceptable' or 'Unacceptable': Attacked prompt: b"Analyze the provided task and call its grammatical correctness as 'Acceptable' or 'Unacceptable':" Original acc: 77.50%, attacked acc: 72.00%, dropped acc: 5.50% Original prompt: Assess the grammatical structure of the given sentence and classify it as 'Acceptable' or 'Unacceptable': Attacked prompt: b"what the grammatical structure of the given sentence and consider it as 'Acceptable' or 'Unacceptable':" Original acc: 78.50%, attacked acc: 0.00%, dropped acc: 78.50% Original prompt: While performing grammar analysis, classify the grammar of the following sentence as 'acceptable' for correct grammar or 'unacceptable' for incorrect grammar: Attacked prompt: b"While performing grammar analysis, classify the grammar of the following sentence as 'acceptable' for so grammar or 'unacceptable' for abnormal grammar:" Original acc: 80.00%, attacked acc: 0.00%, dropped acc: 80.00% Original prompt: Functioning as a grammar evaluation tool, analyze the given sentence and decide if it is grammatically correct, responding with 'acceptable' or 'unacceptable': Attacked prompt: b"also as a grammar analyze assistant, determine the described document and calculate if it is grammatically correct, and with 'acceptable' or 'unacceptable':" Original acc: 78.50%, attacked acc: 0.00%, dropped acc: 78.50% Original prompt: Acting as a grammar detection instrument, determine if the provided sentence is grammatically sound, answering with 'acceptable' for correct grammar or 'unacceptable' for incorrect grammar: Attacked prompt: b"by as a capital detection instrument, detect if the present sentence is grammatically sound, replied with 'acceptable' for correct length or 'unacceptable' for incorrect inclusive:" Original acc: 78.50%, attacked acc: 72.00%, dropped acc: 6.50% ## checklist Original prompt: Review the sentence below and identify whether its grammar is 'Acceptable' or 'Unacceptable': Attacked prompt: b"Review the sentence below and identify whether its grammar is 'Acceptable' or 'Unacceptable' zexKUTWJYG:" Original acc: 78.50%, attacked acc: 72.50%, dropped acc: 6.00% Original prompt: Assess the grammatical structure of the given sentence and classify it as 'Acceptable' or 'Unacceptable': Attacked prompt: b"Assess the grammatical structure of the given sentence and classify it as 'Acceptable' or 'Unacceptable' zexKUTWJYG:" Original acc: 78.00%, attacked acc: 75.50%, dropped acc: 2.50% Original prompt: Analyze the provided sentence and classify its grammatical correctness as 'Acceptable' or 'Unacceptable': Attacked prompt: b"Analyze the provided sentence and classify its grammatical correctness as 'Acceptable' or 'Unacceptable' AMsRIKZniY:" Original acc: 77.50%, attacked acc: 74.00%, dropped acc: 3.50% Original prompt: While performing grammar analysis, classify the grammar of the following sentence as 'acceptable' for correct grammar or 'unacceptable' for incorrect grammar: Attacked prompt: b"While performing grammar analysis, classify the grammar of the following sentence as 'acceptable' for correct grammar or 'unacceptable' for incorrect grammar zexKUTWJYG:" Original acc: 79.00%, attacked acc: 45.00%, dropped acc: 34.00% Original prompt: Acting as a grammar detection instrument, determine if the provided sentence is grammatically sound, answering with 'acceptable' for correct grammar or 'unacceptable' for incorrect grammar: Attacked prompt: b"Acting as a grammar detection instrument, determine if the provided sentence is grammatically sound, answering with 'acceptable' for correct grammar or 'unacceptable' for incorrect grammar RcRneWo6Iv:" Original acc: 79.00%, attacked acc: 73.50%, dropped acc: 5.50% Original prompt: Functioning as a grammar evaluation tool, analyze the given sentence and decide if it is grammatically correct, responding with 'acceptable' or 'unacceptable': Attacked prompt: b"Functioning as a grammar evaluation tool, analyze the given sentence and decide if it is grammatically correct, responding with 'acceptable' or 'unacceptable' IhuBIhoPGc:" Original acc: 79.50%, attacked acc: 73.50%, dropped acc: 6.00% ## deepwordbug Original prompt: Review the sentence below and identify whether its grammar is 'Acceptable' or 'Unacceptable': Attacked prompt: b"Review the sentece below and identify whether its grammar is 'Acceptable' or 'Unacceptable':" Original acc: 79.00%, attacked acc: 74.50%, dropped acc: 4.50% Original prompt: Assess the grammatical structure of the given sentence and classify it as 'Acceptable' or 'Unacceptable': Attacked prompt: b"AIssess the grammatiCcal structure of the given sentednce and classify it as 'Acceptable' or 'Unacceptable':" Original acc: 78.50%, attacked acc: 76.00%, dropped acc: 2.50% Original prompt: Analyze the provided sentence and classify its grammatical correctness as 'Acceptable' or 'Unacceptable': Attacked prompt: b"Analyze the provided sentene and classify its garmmatical correctness as 'Acceptable' or 'Unacceptable':" Original acc: 77.50%, attacked acc: 75.00%, dropped acc: 2.50% Original prompt: While performing grammar analysis, classify the grammar of the following sentence as 'acceptable' for correct grammar or 'unacceptable' for incorrect grammar: Attacked prompt: b"nhile performing grammr analysis, classify the rgammar of the fllowing sentenZe as 'acceptable' for corFect Xrammar or 'unacceptable' for incorrect gramUar:" Original acc: 79.50%, attacked acc: 0.00%, dropped acc: 79.50% Original prompt: Functioning as a grammar evaluation tool, analyze the given sentence and decide if it is grammatically correct, responding with 'acceptable' or 'unacceptable': Attacked prompt: b"Functioning as a grammar evaluation Lool, analyze the Tgiven sentence and deciRe if it is grammXatically correct, respoOnding with 'acceptable' or 'unacceptable':" Original acc: 79.00%, attacked acc: 74.00%, dropped acc: 5.00% Original prompt: Acting as a grammar detection instrument, determine if the provided sentence is grammatically sound, answering with 'acceptable' for correct grammar or 'unacceptable' for incorrect grammar: Attacked prompt: b"cting as a grammar detection instrument, determine if the provided sentence is grammatically sound, aLswering with 'acceptable' for correct grammar or 'unacceptable' for ncorrect gIrammar:" Original acc: 79.00%, attacked acc: 75.00%, dropped acc: 4.00% ## stresstest Original prompt: Review the sentence below and identify whether its grammar is 'Acceptable' or 'Unacceptable': Attacked prompt: b"Review the sentence below and identify whether its grammar is 'Acceptable' or 'Unacceptable' and true is true :" Original acc: 78.50%, attacked acc: 32.50%, dropped acc: 46.00% Original prompt: Assess the grammatical structure of the given sentence and classify it as 'Acceptable' or 'Unacceptable': Attacked prompt: b"Assess the grammatical structure of the given sentence and classify it as 'Acceptable' or 'Unacceptable' and true is true :" Original acc: 78.00%, attacked acc: 59.50%, dropped acc: 18.50% Original prompt: Analyze the provided sentence and classify its grammatical correctness as 'Acceptable' or 'Unacceptable': Attacked prompt: b"Analyze the provided sentence and classify its grammatical correctness as 'Acceptable' or 'Unacceptable' and false is not true :" Original acc: 76.50%, attacked acc: 75.00%, dropped acc: 1.50% Original prompt: Functioning as a grammar evaluation tool, analyze the given sentence and decide if it is grammatically correct, responding with 'acceptable' or 'unacceptable': Attacked prompt: b"Functioning as a grammar evaluation tool, analyze the given sentence and decide if it is grammatically correct, responding with 'acceptable' or 'unacceptable' and false is not true :" Original acc: 79.50%, attacked acc: 76.50%, dropped acc: 3.00% Original prompt: While performing grammar analysis, classify the grammar of the following sentence as 'acceptable' for correct grammar or 'unacceptable' for incorrect grammar: Attacked prompt: b"While performing grammar analysis, classify the grammar of the following sentence as 'acceptable' for correct grammar or 'unacceptable' for incorrect grammar and true is true and true is true and true is true and true is true and true is true :" Original acc: 79.50%, attacked acc: 75.00%, dropped acc: 4.50% Original prompt: Acting as a grammar detection instrument, determine if the provided sentence is grammatically sound, answering with 'acceptable' for correct grammar or 'unacceptable' for incorrect grammar: Attacked prompt: b"Acting as a grammar detection instrument, determine if the provided sentence is grammatically sound, answering with 'acceptable' for correct grammar or 'unacceptable' for incorrect grammar and false is not true :" Original acc: 78.50%, attacked acc: 77.00%, dropped acc: 1.50% ## textbugger Original prompt: Review the sentence below and identify whether its grammar is 'Acceptable' or 'Unacceptable': Attacked prompt: b"Review the punishments below and identify whether its grammatical is 'Acceptable' or 'Unacceptable':" Original acc: 78.50%, attacked acc: 27.50%, dropped acc: 51.00% Original prompt: Assess the grammatical structure of the given sentence and classify it as 'Acceptable' or 'Unacceptable': Attacked prompt: b"As sess the garmmatical structure of the conferred sentence and classify it as 'Acceptable' or 'Unacceptable':" Original acc: 77.50%, attacked acc: 6.50%, dropped acc: 71.00% Original prompt: Analyze the provided sentence and classify its grammatical correctness as 'Acceptable' or 'Unacceptable': Attacked prompt: b"Analyze the furnished sentences and cl\xc9\x91ssify its pronouns correctness as 'Acceptable' or 'Unacceptable':" Original acc: 77.50%, attacked acc: 59.00%, dropped acc: 18.50% Original prompt: While performing grammar analysis, classify the grammar of the following sentence as 'acceptable' for correct grammar or 'unacceptable' for incorrect grammar: Attacked prompt: b"While prforming grammar analyzing, classify the grammar of the follows sent\xd0\xb5nce as 'acceptable' for correcting gammar or 'unacceptable' for incorrect \xc9\xa1rammar:" Original acc: 79.50%, attacked acc: 9.50%, dropped acc: 70.00% Original prompt: Functioning as a grammar evaluation tool, analyze the given sentence and decide if it is grammatically correct, responding with 'acceptable' or 'unacceptable': Attacked prompt: b"Functioning as a gram\xef\xbd\x8dar evaluation instrumentation, analyze the given condemnation and decide if it is grammatically correct, responding with 'acceptable' or 'unacceptable':" Original acc: 78.50%, attacked acc: 74.50%, dropped acc: 4.00% Original prompt: Acting as a grammar detection instrument, determine if the provided sentence is grammatically sound, answering with 'acceptable' for correct grammar or 'unacceptable' for incorrect grammar: Attacked prompt: b"Acting as a grammar detection instrument, dete\xe2\xb2\x85mine if the provided sen\xf0\x9d\x9a\x9dence is grammatially sound, answering with 'acceptable' for correct grammar or 'unacceptable' for incorrect grammar:" Original acc: 79.00%, attacked acc: 73.00%, dropped acc: 6.00% ## textfooler Original prompt: Review the sentence below and identify whether its grammar is 'Acceptable' or 'Unacceptable': Attacked prompt: b"Revising the punishments below and locate whether its dictionary is 'Acceptable' or 'Unacceptable':" Original acc: 79.00%, attacked acc: 2.00%, dropped acc: 77.00% Original prompt: Assess the grammatical structure of the given sentence and classify it as 'Acceptable' or 'Unacceptable': Attacked prompt: b"Assess the pronouns structures of the given convicts and hierarchy it as 'Acceptable' or 'Unacceptable':" Original acc: 78.00%, attacked acc: 17.50%, dropped acc: 60.50% Original prompt: Analyze the provided sentence and classify its grammatical correctness as 'Acceptable' or 'Unacceptable': Attacked prompt: b"Analyse the furnished penalties and classify its grammatical certitude as 'Acceptable' or 'Unacceptable':" Original acc: 77.00%, attacked acc: 21.50%, dropped acc: 55.50% Original prompt: Functioning as a grammar evaluation tool, analyze the given sentence and decide if it is grammatically correct, responding with 'acceptable' or 'unacceptable': Attacked prompt: b"Functioning as a grammar rating toolbox, analyze the given sanctions and decide if it is grammatically correct, responds with 'acceptable' or 'unacceptable':" Original acc: 79.00%, attacked acc: 73.00%, dropped acc: 6.00% Original prompt: While performing grammar analysis, classify the grammar of the following sentence as 'acceptable' for correct grammar or 'unacceptable' for incorrect grammar: Attacked prompt: b"While performing jargon discusses, categories the language of the below sentence as 'acceptable' for mend jargon or 'unacceptable' for unfit grammar:" Original acc: 80.00%, attacked acc: 6.00%, dropped acc: 74.00% Original prompt: Acting as a grammar detection instrument, determine if the provided sentence is grammatically sound, answering with 'acceptable' for correct grammar or 'unacceptable' for incorrect grammar: Attacked prompt: b"Behaving as a jargon detector tool, assess if the tabled punishments is grammatically robust, cater with 'acceptable' for correct grammar or 'unacceptable' for misdirected lexicon:" Original acc: 79.00%, attacked acc: 26.50%, dropped acc: 52.50% ## translation Language: chinese, acc: 79.00%, prompt: As a tool for determining grammatical correctness, please review the following sentences and categorize them grammatically as 'acceptable' or 'unacceptable'. Language: chinese, acc: 77.50%, prompt: Please evaluate the grammar of the following sentences and mark them as 'acceptable' or 'unacceptable'. Language: chinese, acc: 76.50%, prompt: Please check that the following sentences are grammatically correct and categorize them as 'acceptable' or 'unacceptable'. Language: chinese, acc: 76.50%, prompt: Please determine whether the following sentences are 'acceptable' or 'unacceptable' based on grammatical correctness. Language: chinese, acc: 78.50%, prompt: Please categorize the grammar of the following sentences as 'acceptable' or 'unacceptable'. Language: chinese, acc: 78.50%, prompt: Please check the grammar of the following sentences and classify them as 'acceptable' or 'unacceptable'. Language: chinese, acc: 77.50%, prompt: Please mark the following sentences as 'acceptable' or 'unacceptable' according to the grammatical norms. Language: chinese, acc: 77.00%, prompt: Please assess the grammar of the following sentences and classify them as 'acceptable' or 'unacceptable'. Language: chinese, acc: 75.00%, prompt: Please mark the following sentences as 'acceptable' or 'unacceptable' for grammatical accuracy. Language: chinese, acc: 78.50%, prompt: Please check the grammar of the following sentences and classify them as 'acceptable' or 'unacceptable'. Language: french, acc: 72.00%, prompt: As a tool for assessing semantic relevance, categorize the sentence as 'acceptable' or 'unacceptable' according to its meaning and context. Language: french, acc: 75.50%, prompt: Determine whether the grammatical structure of the sentence is 'acceptable' or 'unacceptable' by carefully reviewing it. Language: french, acc: 77.50%, prompt: Judge whether the sentence follows the syntactic rules of the language and classify it as 'acceptable' or 'unacceptable'. Language: french, acc: 69.00%, prompt: Assess the relevance of the sentence by classifying it as 'acceptable' or 'unacceptable' according to the context in which it is used. Language: french, acc: 71.00%, prompt: As a semantic classification tool, decide whether the sentence is 'acceptable' or 'unacceptable' according to its meaning and consistency. Language: french, acc: 70.50%, prompt: Analyze the sentence carefully and determine whether it is grammatically 'acceptable' or 'unacceptable'. Language: french, acc: 74.50%, prompt: As a syntactic validation tool, determine whether the sentence is properly structured and categorize it as 'acceptable' or 'unacceptable'. Language: french, acc: 61.00%, prompt: When evaluating the grammar of the sentence, classify it as 'acceptable' or 'unacceptable' according to the rules of the French language. Language: french, acc: 75.50%, prompt: Determine whether the sentence is 'acceptable' or 'unacceptable' depending on its syntactic construction and semantic context. Language: french, acc: 76.00%, prompt: As a tool for linguistic analysis, categorize the sentence as 'acceptable' or 'unacceptable' according to its relevance and semantic consistency. Language: arabic, acc: 77.50%, prompt: As a tool for determining grammatical correctness, review the sentence and classify its rules as 'acceptable' or 'unacceptable'. Language: arabic, acc: 78.50%, prompt: Analyze the sentence carefully and classify its grammar between 'acceptable' or 'unacceptable'. Language: arabic, acc: 76.00%, prompt: As a tool for determining grammatical correctness, classify the rules of the sentence between 'acceptable' or 'unacceptable'. Language: arabic, acc: 75.50%, prompt: Evaluate the correctness of the sentence between 'acceptable' or 'unacceptable', as a tool for determining grammatical correctness. Language: arabic, acc: 75.00%, prompt: As a tool for classifying valid sentences, determine whether 'acceptable' or 'unacceptable' is based on its grammatical rules. Language: arabic, acc: 76.00%, prompt: Analyze the sentence carefully and classify its grammatical rules between 'acceptable' or 'unacceptable'. Language: arabic, acc: 76.00%, prompt: Analyze the sentence and classify it between 'acceptable' or 'unacceptable' as a grammatical check tool. Language: arabic, acc: 76.00%, prompt: As a classification tool for grammatical sentences, determine whether the sentence 'acceptable' or 'unacceptable' is based on its rules. Language: arabic, acc: 72.00%, prompt: As a tool for analyzing grammar, classify the rules of the sentence between 'acceptable' or 'unacceptable'. Language: arabic, acc: 75.00%, prompt: Determine whether the sentence is 'acceptable' or 'unacceptable' as a tool for determining grammatical correctness and analyze its grammar. Language: spanish, acc: 77.00%, prompt: As a tool to determine grammatical correctness, review the sentence and categorize its grammar as 'acceptable' or 'unacceptable'. Language: spanish, acc: 76.50%, prompt: Analyze the sentence carefully and classify its grammar as 'acceptable' or 'unacceptable'. Language: spanish, acc: 75.50%, prompt: As a tool for determining grammatical correctness, it categorizes the grammatical rules of the sentence as 'acceptable' or 'unacceptable'. Language: spanish, acc: 75.50%, prompt: Evaluate the grammatical correctness of the sentence and classify it as 'acceptable' or 'unacceptable' using a grammatical verification tool. Language: spanish, acc: 76.50%, prompt: As a tool for classifying grammatically correct sentences, determine whether the sentence is 'acceptable' or 'unacceptable' based on its grammatical rules. Language: spanish, acc: 74.00%, prompt: Analyse the sentence carefully and classify its grammar as 'acceptable' or 'unacceptable' using a grammatical revision tool. Language: spanish, acc: 77.00%, prompt: As a tool for classifying grammatical sentences, it determines whether the sentence is 'acceptable' or 'unacceptable' using its grammatical structure. Language: spanish, acc: 75.00%, prompt: As a tool for analyzing grammatical correctness, it categorizes the grammatical rules of the sentence as 'acceptable' or 'unacceptable'. Language: spanish, acc: 76.50%, prompt: Determine whether the sentence is 'acceptable' or 'unacceptable' using a grammatical verification tool and sort it accordingly. Language: spanish, acc: 75.50%, prompt: As a tool for determining grammatical correctness, evaluate the sentence and classify it as 'acceptable' or 'unacceptable' based on its grammatical rules. Language: japanese, acc: 75.50%, prompt: As a tool to determine whether grammar is grammatically correct, look at the sentence and categorize grammar into the 'acceptable' or 'unacceptable' categories. Language: japanese, acc: 78.00%, prompt: Please read the given sentence and categorize the grammar into the 'acceptable' or 'unacceptable' categories. Language: japanese, acc: 75.00%, prompt: To determine the correctness of a sentence, categorize grammar into two categories: 'acceptable' or 'unacceptable'. Language: japanese, acc: 76.00%, prompt: Classify the grammar of a given sentence into two categories: 'acceptable' and 'unacceptable'. Language: japanese, acc: 79.00%, prompt: Classify a given sentence into the categories 'acceptable' or 'unacceptable' to determine whether it is grammatically correct. Language: japanese, acc: 77.00%, prompt: To determine whether it is grammatically correct, categorize a given sentence into two categories: 'acceptable' or 'unacceptable'. Language: japanese, acc: 77.00%, prompt: To determine the correctness of grammar, categorize a given sentence into two categories: 'acceptable' or 'unacceptable'. Language: japanese, acc: 25.00%, prompt: Classify the grammar of a given sentence into two categories, 'acceptable' or 'unacceptable', and judge its accuracy. Language: japanese, acc: 77.50%, prompt: To determine whether it is grammatically correct, divide a given sentence into two categories: 'acceptable' or 'unacceptable'. Language: japanese, acc: 76.50%, prompt: To evaluate the accuracy of grammar, categorize a given sentence into two categories: 'acceptable' or 'unacceptable'. Language: korean, acc: 76.50%, prompt: As a tool for judging grammatical correctness, please review the sentences and classify the grammar as 'acceptable' or 'unacceptable'. Language: korean, acc: 75.50%, prompt: Please read the given sentences, and classify the grammar as 'acceptable' or 'unacceptable'. Language: korean, acc: 78.00%, prompt: Please classify the sentences as 'acceptable' or 'unacceptable' to judge the grammatical appropriateness. Language: korean, acc: 75.50%, prompt: Please classify the grammar of the given sentences into 2 categories: 'acceptable' or 'unacceptable'. Language: korean, acc: 76.50%, prompt: Please classify the given sentences as 'acceptable' or 'unacceptable' to judge the correctness of the grammar. Language: korean, acc: 75.50%, prompt: Please classify the given sentences as 'acceptable' or 'unacceptable' in order to judge their grammatical appropriateness. Language: korean, acc: 76.50%, prompt: Please determine the correctness of your grammar by classifying sentences as 'acceptable' or 'unacceptable'. Language: korean, acc: 39.50%, prompt: Classify the grammar of a given sentence as 'acceptable' or 'unacceptable', and judge its accuracy. Language: korean, acc: 76.50%, prompt: Please classify the given sentences as 'acceptable' or 'unacceptable' to judge the correctness of the grammar. Language: korean, acc: 72.50%, prompt: Please rate the accuracy of your grammar by categorizing sentences as 'acceptable' or 'unacceptable'. # mnli ## 10 prompts Acc: 80.00%, prompt: Assess the connection between the following sentences and classify it as 'entailment', 'neutral', or 'contradiction': Acc: 79.50%, prompt: Considering the two sentences, identify if their relationship is 'entailment', 'neutral', or 'contradiction': Acc: 79.00%, prompt: Analyze the two provided sentences and decide if their relationship is 'entailment', 'neutral', or 'contradiction': Acc: 79.00%, prompt: Identify whether the given pair of sentences demonstrates entailment, neutral, or contradiction. Answer with 'entailment', 'neutral', or 'contradiction': Acc: 79.00%, prompt: Review the two statements and categorize their relationship as either 'entailment', 'neutral', or 'contradiction': Acc: 78.50%, prompt: Examine the pair of sentences and determine if they exhibit entailment, neutral, or contradiction. Answer with either 'entailment', 'neutral', or 'contradiction': Acc: 77.50%, prompt: Please classify the relationship between the provided sentences as 'entailment', 'neutral', or 'contradiction': Acc: 77.50%, prompt: Indicate if the connection between the following sentences is 'entailment', 'neutral', or 'contradiction': Acc: 77.50%, prompt: Determine if the given pair of sentences displays entailment, neutral, or contradiction. Respond with 'entailment', 'neutral', or 'contradiction': Acc: 76.50%, prompt: Does the relationship between the given sentences represent entailment, neutral, or contradiction? Respond with 'entailment', 'neutral', or 'contradiction': Acc: 81.00%, prompt: In the capacity of an entailment assessment system, indicate if the link between the following sentences is 'entailment', 'neutral', or 'contradiction': Acc: 80.00%, prompt: As an instrument for entailment evaluation, consider the two sentences and determine if their relationship is 'entailment', 'neutral', or 'contradiction': Acc: 79.50%, prompt: In the role of a semantic relationship analyzer, examine the connection between the given sentences and decide if they exhibit entailment, neutral, or contradiction. Answer with 'entailment', 'neutral', or 'contradiction': Acc: 79.00%, prompt: As a tool for determining entailment relationships, review the two statements and categorize their connection as either 'entailment', 'neutral', or 'contradiction': Acc: 78.50%, prompt: Functioning as an entailment evaluation tool, analyze the provided sentences and decide if their relationship is 'entailment', 'neutral', or 'contradiction': Acc: 78.00%, prompt: In your role as an entailment analysis tool, assess the relationship between the given sentences and classify it as 'entailment', 'neutral', or 'contradiction': Acc: 77.50%, prompt: As an entailment identification system, examine the connection between the following sentences and respond with 'entailment', 'neutral', or 'contradiction': Acc: 77.50%, prompt: Acting as an entailment detection instrument, determine if the given pair of sentences demonstrates entailment, neutral, or contradiction. Answer with 'entailment', 'neutral', or 'contradiction': Acc: 77.50%, prompt: Working as an entailment classifier, identify whether the given pair of sentences displays entailment, neutral, or contradiction. Respond with 'entailment', 'neutral', or 'contradiction': Acc: 77.00%, prompt: While performing entailment analysis, classify the relationship between the provided sentences as 'entailment', 'neutral', or 'contradiction': ## bertattack Original prompt: Assess the connection between the following sentences and classify it as 'entailment', 'neutral', or 'contradiction': Attacked prompt: b"do the reconciliation between the following sentences and distinguish it as 'entailment', 'neutral', or 'contradiction':" Original acc: 79.00%, attacked acc: 61.00%, dropped acc: 18.00% Original prompt: Considering the two sentences, identify if their relationship is 'entailment', 'neutral', or 'contradiction': Attacked prompt: b"Considering the both sentences, discuss if their love is 'entailment', 'neutral', or 'contradiction':" Original acc: 78.50%, attacked acc: 39.50%, dropped acc: 39.00% Original prompt: Analyze the two provided sentences and decide if their relationship is 'entailment', 'neutral', or 'contradiction': Attacked prompt: b"Analyze the two provided sentences and read if their sex is 'entailment', 'neutral', or 'contradiction':" Original acc: 79.00%, attacked acc: 69.00%, dropped acc: 10.00% Original prompt: In the capacity of an entailment assessment system, indicate if the link between the following sentences is 'entailment', 'neutral', or 'contradiction': Attacked prompt: b"what the significance of an entailment assessment system, be if the link between the following stimuli is 'entailment', 'neutral', or 'contradiction':" Original acc: 81.00%, attacked acc: 13.00%, dropped acc: 68.00% Original prompt: As an instrument for entailment evaluation, consider the two sentences and determine if their relationship is 'entailment', 'neutral', or 'contradiction': Attacked prompt: b"As an instrument for entailment assessing, know the two proposition and prove if their contradiction is 'entailment', 'neutral', or 'contradiction':" Original acc: 78.50%, attacked acc: 44.50%, dropped acc: 34.00% Original prompt: In the role of a semantic relationship analyzer, examine the connection between the given sentences and decide if they exhibit entailment, neutral, or contradiction. Answer with 'entailment', 'neutral', or 'contradiction': Attacked prompt: b"at the cost of a semantics task analyze!, examine the tension between the given persons and decide if they exhibit entailment, neutral, or contradiction. open with 'entailment', 'neutral', or 'contradiction':" Original acc: 79.50%, attacked acc: 62.00%, dropped acc: 17.50% ## checklist Original prompt: Working as an entailment classifier, identify whether the given pair of sentences displays entailment, neutral, or contradiction. Respond with 'entailment', 'neutral', or 'contradiction'. Attacked prompt: b"Working as an entailment classifier, identify whether the given pair of sentences displays entailment, neutral, or contradiction. Respond with 'entailment', 'neutral', or 'contradiction' 9vGXgnbWB8. " Original acc: 87.90%, attacked acc: 87.90%, dropped acc: 0.00% Original prompt: Considering the two sentences, identify if their relationship is 'entailment', 'neutral', or 'contradiction': Attacked prompt: b"Considering the two sentences, identify if their relationship is 'entailment', 'neutral', or 'contradiction' 5XeflW1ZJc:" Original acc: 80.00%, attacked acc: 73.50%, dropped acc: 6.50% Original prompt: Analyze the two provided sentences and decide if their relationship is 'entailment', 'neutral', or 'contradiction': Attacked prompt: b"Analyze the two provided sentences and decide if their relationship is 'entailment', 'neutral', or 'contradiction' 5yWbBXztUY:" Original acc: 79.50%, attacked acc: 74.00%, dropped acc: 5.50% Original prompt: In the capacity of an entailment assessment system, indicate if the link between the following sentences is 'entailment', 'neutral', or 'contradiction': Attacked prompt: b"In the capacity of an entailment assessment system, indicate if the link between the following sentences is 'entailment', 'neutral', or 'contradiction' vTAjHynoIG:" Original acc: 80.50%, attacked acc: 75.50%, dropped acc: 5.00% Original prompt: In the role of a semantic relationship analyzer, examine the connection between the given sentences and decide if they exhibit entailment, neutral, or contradiction. Answer with 'entailment', 'neutral', or 'contradiction': Attacked prompt: b"In the role of a semantic relationship analyzer, examine the connection between the given sentences and decide if they exhibit entailment, neutral, or contradiction. Answer with 'entailment', 'neutral', or 'contradiction' g0vBZf3tQC:" Original acc: 79.00%, attacked acc: 75.00%, dropped acc: 4.00% Original prompt: As an instrument for entailment evaluation, consider the two sentences and determine if their relationship is 'entailment', 'neutral', or 'contradiction': Attacked prompt: b"As an instrument for entailment evaluation, consider the two sentences and determine if their relationship is 'entailment', 'neutral', or 'contradiction' 5yWbBXztUY:" Original acc: 80.00%, attacked acc: 76.00%, dropped acc: 4.00% ## deepwordbug Original prompt: Assess the connection between the following sentences and classify it as 'entailment', 'neutral', or 'contradiction': Attacked prompt: b"Asess the cnonection between the following sentences and classify it as 'entailment', 'neutral', or 'contradiction':" Original acc: 79.50%, attacked acc: 72.00%, dropped acc: 7.50% Original prompt: Analyze the two provided sentences and decide if their relationship is 'entailment', 'neutral', or 'contradiction': Attacked prompt: b"Analyze the two porvided sentences and decidle if their relationKhip is 'entailment', 'neutral', or 'contradiction':" Original acc: 79.00%, attacked acc: 75.00%, dropped acc: 4.00% Original prompt: Considering the two sentences, identify if their relationship is 'entailment', 'neutral', or 'contradiction': Attacked prompt: b"Aonsidering the twH sentences, identifky if their reltionship is 'entailment', 'neutral', or 'contradiction':" Original acc: 79.50%, attacked acc: 75.50%, dropped acc: 4.00% Original prompt: In the capacity of an entailment assessment system, indicate if the link between the following sentences is 'entailment', 'neutral', or 'contradiction': Attacked prompt: b"In the capacity of an entailment assessment syste, indiNate if the link between the following rsentences is 'entailment', 'neutral', or 'contradiction':" Original acc: 80.50%, attacked acc: 75.00%, dropped acc: 5.50% Original prompt: In the role of a semantic relationship analyzer, examine the connection between the given sentences and decide if they exhibit entailment, neutral, or contradiction. Answer with 'entailment', 'neutral', or 'contradiction': Attacked prompt: b"In the role of a semantXc relationship analyzer, examine the connetion between the given sentences and decide if they exhibit entailment, neutral, or contradiction. Answer with 'entailment', 'neutral', or 'contradiction':" Original acc: 79.00%, attacked acc: 76.00%, dropped acc: 3.00% Original prompt: As an instrument for entailment evaluation, consider the two sentences and determine if their relationship is 'entailment', 'neutral', or 'contradiction': Attacked prompt: b"As an instrument for entailment valuation, consider the two sentences and Fdetermine if their relaitonship is 'entailment', 'neutral', or 'contradiction':" Original acc: 78.50%, attacked acc: 72.00%, dropped acc: 6.50% ## stresstest Original prompt: Considering the two sentences, identify if their relationship is 'entailment', 'neutral', or 'contradiction': Attacked prompt: b"Considering the two sentences, identify if their relationship is 'entailment', 'neutral', or 'contradiction' and false is not true :" Original acc: 79.00%, attacked acc: 76.00%, dropped acc: 3.00% Original prompt: Assess the connection between the following sentences and classify it as 'entailment', 'neutral', or 'contradiction': Attacked prompt: b"Assess the connection between the following sentences and classify it as 'entailment', 'neutral', or 'contradiction' and true is true :" Original acc: 79.50%, attacked acc: 74.50%, dropped acc: 5.00% Original prompt: Analyze the two provided sentences and decide if their relationship is 'entailment', 'neutral', or 'contradiction': Attacked prompt: b"Analyze the two provided sentences and decide if their relationship is 'entailment', 'neutral', or 'contradiction' and false is not true :" Original acc: 79.50%, attacked acc: 76.00%, dropped acc: 3.50% Original prompt: In the capacity of an entailment assessment system, indicate if the link between the following sentences is 'entailment', 'neutral', or 'contradiction': Attacked prompt: b"In the capacity of an entailment assessment system, indicate if the link between the following sentences is 'entailment', 'neutral', or 'contradiction' and true is true and true is true and true is true and true is true and true is true :" Original acc: 80.50%, attacked acc: 78.50%, dropped acc: 2.00% Original prompt: As an instrument for entailment evaluation, consider the two sentences and determine if their relationship is 'entailment', 'neutral', or 'contradiction': Attacked prompt: b"As an instrument for entailment evaluation, consider the two sentences and determine if their relationship is 'entailment', 'neutral', or 'contradiction' and false is not true :" Original acc: 80.00%, attacked acc: 78.50%, dropped acc: 1.50% Original prompt: In the role of a semantic relationship analyzer, examine the connection between the given sentences and decide if they exhibit entailment, neutral, or contradiction. Answer with 'entailment', 'neutral', or 'contradiction': Attacked prompt: b"In the role of a semantic relationship analyzer, examine the connection between the given sentences and decide if they exhibit entailment, neutral, or contradiction. Answer with 'entailment', 'neutral', or 'contradiction' and true is true and true is true and true is true and true is true and true is true :" Original acc: 78.50%, attacked acc: 78.50%, dropped acc: 0.00% ## textbugger Original prompt: Considering the two sentences, identify if their relationship is 'entailment', 'neutral', or 'contradiction': Attacked prompt: b"Considering the ii sentences, identify if their rela\xf0\x9d\x9a\x9dionship is 'entailment', 'neutral', or 'contradiction':" Original acc: 80.00%, attacked acc: 73.50%, dropped acc: 6.50% Original prompt: Assess the connection between the following sentences and classify it as 'entailment', 'neutral', or 'contradiction': Attacked prompt: b"Assses the connecting between the following sentence and cla\xd1\x95sify it as 'entailment', 'neutral', or 'contradiction':" Original acc: 79.50%, attacked acc: 67.50%, dropped acc: 12.00% Original prompt: Analyze the two provided sentences and decide if their relationship is 'entailment', 'neutral', or 'contradiction': Attacked prompt: b"Analyze the two pr\xd0\xbevided punishments and decide if their relatinship is 'entailment', 'neutral', or 'contradiction':" Original acc: 78.50%, attacked acc: 46.50%, dropped acc: 32.00% Original prompt: As an instrument for entailment evaluation, consider the two sentences and determine if their relationship is 'entailment', 'neutral', or 'contradiction': Attacked prompt: b"Iike an instrument for entailment eval\xd5\xbdation, consider the two verdicts and determine if their relatio\xd5\xb8ship is 'entailment', 'neutral', or 'contradiction':" Original acc: 79.50%, attacked acc: 70.00%, dropped acc: 9.50% Original prompt: In the capacity of an entailment assessment system, indicate if the link between the following sentences is 'entailment', 'neutral', or 'contradiction': Attacked prompt: b"In the capacity of an entailment assessment system, depicts if the link between the subsequently verdicts is 'entailment', 'neutral', or 'contradiction':" Original acc: 81.00%, attacked acc: 69.50%, dropped acc: 11.50% Original prompt: In the role of a semantic relationship analyzer, examine the connection between the given sentences and decide if they exhibit entailment, neutral, or contradiction. Answer with 'entailment', 'neutral', or 'contradiction': Attacked prompt: b"In the role of a sem antic relationships analyze\xe2\xb2\x85, examine the connection between the given punishments and decide if they \xd0\xb5xhibit entailment, neutral, or contradiction. Answers with 'entailment', 'neutral', or 'contradiction':" Original acc: 79.00%, attacked acc: 62.00%, dropped acc: 17.00% ## textfooler Original prompt: Assess the connection between the following sentences and classify it as 'entailment', 'neutral', or 'contradiction': Attacked prompt: b"Assess the liaison between the next convicting and classify it as 'entailment', 'neutral', or 'contradiction':" Original acc: 80.00%, attacked acc: 59.00%, dropped acc: 21.00% Original prompt: Considering the two sentences, identify if their relationship is 'entailment', 'neutral', or 'contradiction': Attacked prompt: b"Discuss the two punishment, discovering if their coitus is 'entailment', 'neutral', or 'contradiction':" Original acc: 78.50%, attacked acc: 10.50%, dropped acc: 68.00% Original prompt: Analyze the two provided sentences and decide if their relationship is 'entailment', 'neutral', or 'contradiction': Attacked prompt: b"Explore the two offered punishments and choices if their coitus is 'entailment', 'neutral', or 'contradiction':" Original acc: 78.50%, attacked acc: 4.50%, dropped acc: 74.00% Original prompt: In the capacity of an entailment assessment system, indicate if the link between the following sentences is 'entailment', 'neutral', or 'contradiction': Attacked prompt: b"Through the jurisdiction of an entailment review mechanism, explain if the binding between the following punishments is 'entailment', 'neutral', or 'contradiction':" Original acc: 81.00%, attacked acc: 31.00%, dropped acc: 50.00% Original prompt: As an instrument for entailment evaluation, consider the two sentences and determine if their relationship is 'entailment', 'neutral', or 'contradiction': Attacked prompt: b"As an instrument for entailment reviewed, discuss the two verdicts and deciding if their coitus is 'entailment', 'neutral', or 'contradiction':" Original acc: 80.00%, attacked acc: 51.50%, dropped acc: 28.50% Original prompt: In the role of a semantic relationship analyzer, examine the connection between the given sentences and decide if they exhibit entailment, neutral, or contradiction. Answer with 'entailment', 'neutral', or 'contradiction': Attacked prompt: b"De the role of a semantic connection analyzer, researched the connection between the given sorrows and ascertain if they illustrates entailment, neutral, or contradiction. Rebuttal with 'entailment', 'neutral', or 'contradiction':" Original acc: 78.50%, attacked acc: 57.00%, dropped acc: 21.50% ## translation Language: chinese, acc: 76.50%, prompt: As an implication analysis tool, evaluate the relationships between given sentences and classify them as 'entailment', 'neutral', or 'contradiction'. Language: chinese, acc: 76.00%, prompt: Analyze the relationships between the given sentences and classify them as 'entailment', 'neutral', or 'contradiction'. Language: chinese, acc: 78.00%, prompt: Evaluate the logical relationship between the following sentences and classify them as 'entailment', 'neutral', or 'contradiction'. Language: chinese, acc: 78.00%, prompt: In your role as an implication analysis tool, judge the relationships of the following sentences and classify them as 'entailment', 'neutral', or 'contradiction'. Language: chinese, acc: 80.00%, prompt: Please judge the relationship between the following sentences as 'entailment', 'neutral', or 'contradiction'. Language: chinese, acc: 73.50%, prompt: From a given sentence, evaluate the relationship between them and classify them as 'entailment', 'neutral', or 'contradiction'. Language: chinese, acc: 75.00%, prompt: Please classify the following sentences as 'entailment', 'neutral', or 'contradiction' according to their relationships. Language: chinese, acc: 76.50%, prompt: Evaluate the logical relationship between the following sentences and classify them as 'entailment', 'neutral', or 'contradiction'. Language: chinese, acc: 76.00%, prompt: As an implication analysis tool, categorize the relationships between given sentences as 'entailment', 'neutral', or 'contradiction'. Language: chinese, acc: 74.00%, prompt: Please classify the following sentences as 'entailment', 'neutral', or 'contradiction' according to their logical relationship. Language: french, acc: 76.00%, prompt: As a tool for analyzing the consequence relationship, evaluate the relationship between the given sentences and classify it as 'entailment', 'neutral', or 'contradiction'. Language: french, acc: 76.50%, prompt: Evaluate the relationship between the given sentences and classify it as 'entailment', 'neutral', or 'contradiction'. Language: french, acc: 74.00%, prompt: Determine whether the following sentences are related to 'entailment', 'neutral', or 'contradiction'. Language: french, acc: 73.50%, prompt: In your role as a consequence analysis tool, evaluate the relationship between the given sentences and classify it as 'entailment', 'neutral', or 'contradiction'. Language: french, acc: 77.00%, prompt: Classify the relationship between the following sentences as 'entailment', 'neutral', or 'contradiction'. Language: french, acc: 75.00%, prompt: As a consequence analysis tool, evaluate the relationship between the given sentences and classify it as 'entailment', 'neutral', or 'contradiction'. Language: french, acc: 72.00%, prompt: Analyze the relationship between the given sentences and determine whether it is of 'entailment', 'neutral', or 'contradiction'. Language: french, acc: 76.50%, prompt: Evaluate the relationship between the following sentences and classify it as 'entailment', 'neutral', or 'contradiction'. Language: french, acc: 77.50%, prompt: As a tool for analyzing the consequence relationship, classify the following sentences as 'entailment', 'neutral', or 'contradiction'. Language: french, acc: 72.50%, prompt: Determine whether the given sentences are related to 'entailment', 'neutral', or 'contradiction'. Language: arabic, acc: 70.50%, prompt: Based on your role as a reasoning analyst, analyze the relationship between the given sentences and classify them as 'entailment', 'neutral', or 'contradiction'. Language: arabic, acc: 77.50%, prompt: Evaluate the relationship between given sentences and classify them as 'entailment', 'neutral', or 'contradiction'. Language: arabic, acc: 77.00%, prompt: Determine if the following sentences are 'entailment', 'neutral', or 'contradiction'. Language: arabic, acc: 73.50%, prompt: In your role as a tool of reasoning analysis, investigate the relationship between sentences and classify them as 'entailment', 'neutral', or 'contradiction'. Language: arabic, acc: 77.00%, prompt: Classify the relationship between the following sentences as 'entailment', 'neutral', or 'contradiction'. Language: arabic, acc: 71.50%, prompt: In your role as a tool of reasoning analysis, evaluate the relationship between the given sentences and classify them as 'entailment', 'neutral', or 'contradiction'. Language: arabic, acc: 78.00%, prompt: Analyze the relationship between the given sentences and determine if they are 'entailment', 'neutral', or 'contradiction'. Language: arabic, acc: 76.50%, prompt: Evaluate the relationship between the following sentences and classify them as 'entailment', 'neutral', or 'contradiction'. Language: arabic, acc: 77.00%, prompt: In your role as a tool of reasoning analysis, the following sentences are classified as 'entailment', 'neutral', or 'contradiction'. Language: arabic, acc: 75.00%, prompt: Determine if the sentences given are 'entailment', 'neutral', or 'contradiction'. Language: spanish, acc: 75.50%, prompt: In your role as an implication analysis tool, evaluate the relationship between the given phrases and classify them as 'entailment', 'neutral', or 'contradiction'. Language: spanish, acc: 75.50%, prompt: Determine whether there is 'entailment', 'neutral', or 'contradiction' between the sentences given, using this text analysis tool, Language: spanish, acc: 74.50%, prompt: Analyze the relationship between the two sentences and classify it as 'entailment', 'neutral', or 'contradiction' using this text classification tool, Language: spanish, acc: 80.00%, prompt: Using this implication analysis tool, decide whether the sentences given are related by 'entailment', 'neutral', or 'contradiction'. Language: spanish, acc: 73.00%, prompt: Classifies the relationship between the given phrases as 'entailment', 'neutral', or 'contradiction' using this text analysis tool, Language: spanish, acc: 72.50%, prompt: Evaluate whether there is 'entailment', 'neutral', or 'contradiction' between the sentences provided using this text classification tool, Language: spanish, acc: 76.00%, prompt: Using this implication analysis tool, decide whether the two sentences are related by 'entailment', 'neutral', or 'contradiction'. Language: spanish, acc: 76.00%, prompt: Determine whether the given phrases are related by 'entailment', 'neutral', or 'contradiction' using this text analysis tool, Language: spanish, acc: 77.00%, prompt: Analyze the relationship between the two sentences and classify it as 'entailment', 'neutral', or 'contradiction' using this text analysis tool, Language: spanish, acc: 70.50%, prompt: Using this text classification tool, it classifies the relationship between the given phrases as 'entailment', 'neutral', or 'contradiction'. Language: japanese, acc: 77.50%, prompt: As your role as an implication analysis tool, evaluate the relationship of a given sentence and classify it as 'entailment', 'neutral', or 'contradiction'. Language: japanese, acc: 78.50%, prompt: Use the implication analysis tool as your role to evaluate the relationship of a given sentence and classify it as 'entailment', 'neutral', or 'contradiction'. Language: japanese, acc: 71.00%, prompt: Use this text classification tool to categorize relationships in a given text as 'entailment', 'neutral', or 'contradiction'. Language: japanese, acc: 78.00%, prompt: Use the implication analysis tool as your role and classify the relationship of a given sentence as 'entailment', 'neutral', or 'contradiction'. Language: japanese, acc: 71.50%, prompt: Evaluate the relationship of a given sentence and use this text classification tool to classify it as 'entailment', 'neutral', or 'contradiction'. Language: japanese, acc: 71.00%, prompt: Evaluate the relationship of a given sentence and use this text classification tool to accurately classify it as 'entailment', 'neutral', or 'contradiction'. Language: japanese, acc: 75.50%, prompt: Use the implication analysis tool as your role and use this text classification tool to classify the relationship of a given sentence as 'entailment', 'neutral', or 'contradiction'. Language: japanese, acc: 72.50%, prompt: Use this text classification tool to evaluate the relationship of a given sentence and classify it as 'entailment', 'neutral', or 'contradiction'. Language: japanese, acc: 75.00%, prompt: Use the implication analysis tool as your role, evaluate the relationship of a given sentence, and use this text classification tool to classify it as 'entailment', 'neutral', or 'contradiction'. Language: japanese, acc: 76.50%, prompt: Use the implication analysis tool as your role and categorize the relationship of a given sentence strictly as 'entailment', 'neutral', or 'contradiction' using this text classification tool. Language: korean, acc: 75.50%, prompt: Analyze the relationships between given sentences and classify them as 'entailment', 'neutral', or 'contradiction'. Language: korean, acc: 77.50%, prompt: In the text categorization task, identify the relationship between given sentences as one of 'entailment', 'neutral', or 'contradiction'. Language: korean, acc: 76.00%, prompt: Perform the role of analyzing the relationship between sentences and classifying them as 'entailment', 'neutral', or 'contradiction'. Language: korean, acc: 80.00%, prompt: Evaluate the relationship between two given sentences, and classify them as 'entailment', 'neutral', or 'contradiction'. Language: korean, acc: 73.00%, prompt: In the text categorization task, perform the role of classifying relationships between given sentences as 'entailment', 'neutral', or 'contradiction'. Language: korean, acc: 75.00%, prompt: Judge the associations between sentences, and classify them as 'entailment', 'neutral', or 'contradiction'. Language: korean, acc: 78.00%, prompt: Analyze the relationship between two given sentences and classify them as 'entailment', 'neutral', or 'contradiction'. Language: korean, acc: 71.00%, prompt: In the task of text classification, identify the relationships between given sentences as 'entailment', 'neutral', or 'contradiction'. Language: korean, acc: 73.50%, prompt: Act as an instrument to evaluate the relationships between sentences, and classify them as 'entailment', 'neutral', or 'contradiction'. Language: korean, acc: 78.50%, prompt: Analyze the associations of two given sentences and classify them as 'entailment', 'neutral', or 'contradiction'. # mrpc ## 10 prompts Acc: 76.50%, prompt: Can the given sentences be considered semantically identical? Please reply with 'equivalent' or 'not_equivalent'. Acc: 76.00%, prompt: Examine the pair of sentences and decide if their meanings are identical by answering with 'equivalent' or 'not_equivalent'. Acc: 75.50%, prompt: Evaluate whether the two provided sentences convey the same meaning by answering 'equivalent' or 'not_equivalent'. Acc: 75.00%, prompt: Please analyze the provided sentences and indicate if their meanings are the same by choosing 'equivalent' or 'not_equivalent'. Acc: 74.50%, prompt: Compare the following sentences and determine if they share the same semantic meaning by responding with 'equivalent' or 'not_equivalent'. Acc: 74.50%, prompt: Determine if the meanings of the following sentences are semantically equivalent by responding with 'equivalent' or 'not_equivalent'. Acc: 73.50%, prompt: Do the meanings of these two statements align? Indicate your answer with 'equivalent' or 'not_equivalent'. Acc: 73.00%, prompt: Do these two sentences have the same underlying meaning? Respond with 'equivalent' or 'not_equivalent'. Acc: 73.00%, prompt: Assess if the two given sentences have equivalent meanings by selecting 'equivalent' or 'not_equivalent'. Acc: 71.50%, prompt: Are the meanings of the following pair of sentences the same? Answer with 'equivalent' or 'not_equivalent'. Acc: 76.00%, prompt: In your capacity as a semantic assessment tool, evaluate the provided sentences and classify their meanings as 'equivalent' or 'not_equivalent'. Acc: 75.50%, prompt: As a sentence similarity evaluator, analyze the provided sentences and indicate if their meanings are 'equivalent' or 'not_equivalent'. Acc: 75.50%, prompt: As a linguistic comparator, review the following pair of sentences and determine their semantic equivalence by choosing 'equivalent' or 'not_equivalent'. Acc: 75.50%, prompt: In the role of a sentence comparison analyst, assess the provided sentences and indicate if they convey the same meaning by selecting 'equivalent' or 'not_equivalent'. Acc: 75.50%, prompt: As a textual similarity evaluator, analyze the following pair of sentences and determine if they are semantically 'equivalent' or 'not_equivalent'. Acc: 75.00%, prompt: As a semantic comparison expert, evaluate the given pair of sentences and determine if they are 'equivalent' or 'not_equivalent'. Acc: 75.00%, prompt: In the role of a textual comparison specialist, examine the given sentences and decide if they share the same meaning, responding with 'equivalent' or 'not_equivalent'. Acc: 74.50%, prompt: In your capacity as a semantic comparison tool, examine the given sentences and decide if their meanings are identical, responding with 'equivalent' or 'not_equivalent'. Acc: 73.50%, prompt: As a language comparison expert, examine the given pair of sentences and decide if their meanings align, answering with 'equivalent' or 'not_equivalent'. Acc: 71.00%, prompt: In your capacity as a language analyst, assess the following sentences and classify their similarity as 'equivalent' or 'not_equivalent'. ## bertattack Original prompt: Can the given sentences be considered semantically identical? Please reply with 'equivalent' or 'not_equivalent'. Attacked prompt: b"any the attached words be considered semantically identical? Please reply with 'equivalent' or 'not_equivalent'. " Original acc: 77.00%, attacked acc: 71.00%, dropped acc: 6.00% Original prompt: Evaluate whether the two provided sentences convey the same meaning by answering 'equivalent' or 'not_equivalent'. Attacked prompt: b"we do the two present sentences convey the same meaning by answering 'equivalent' or 'not_equivalent'. " Original acc: 75.00%, attacked acc: 69.50%, dropped acc: 5.50% Original prompt: Examine the pair of sentences and decide if their meanings are identical by answering with 'equivalent' or 'not_equivalent'. Attacked prompt: b"Examine the pairs of pronouns and evaluate if their meanings are different by answering with 'equivalent' or 'not_equivalent'. " Original acc: 75.50%, attacked acc: 56.00%, dropped acc: 19.50% Original prompt: In your capacity as a semantic assessment tool, evaluate the provided sentences and classify their meanings as 'equivalent' or 'not_equivalent'. Attacked prompt: b"In your capacity as a semantic assessment tool, evaluate the provided tense and classify their constituents as 'equivalent' or 'not_equivalent'. " Original acc: 75.50%, attacked acc: 0.00%, dropped acc: 75.50% Original prompt: As a sentence similarity evaluator, analyze the provided sentences and indicate if their meanings are 'equivalent' or 'not_equivalent'. Attacked prompt: b"create a sentence similarity evaluator, contrast the and meanings and indicate if their meanings are 'equivalent' or 'not_equivalent'. " Original acc: 75.50%, attacked acc: 0.00%, dropped acc: 75.50% Original prompt: As a linguistic comparator, review the following pair of sentences and determine their semantic equivalence by choosing 'equivalent' or 'not_equivalent'. Attacked prompt: b"as a linguistic comparator, review the following set of goods and locate their linguistic differences by comparing 'equivalent' or 'not_equivalent'. " Original acc: 75.50%, attacked acc: 39.00%, dropped acc: 36.50% ## checklist Original prompt: Can the given sentences be considered semantically identical? Please reply with 'equivalent' or 'not_equivalent'. Attacked prompt: b"Can the given sentences be considered semantically identical? Please reply with 'equivalent' or 'not_equivalent' MQnugHcaoy. " Original acc: 78.00%, attacked acc: 72.50%, dropped acc: 5.50% Original prompt: Examine the pair of sentences and decide if their meanings are identical by answering with 'equivalent' or 'not_equivalent'. Attacked prompt: b"Examine the pair of sentences and decide if their meanings are identical by answering with 'equivalent' or 'not_equivalent' SFPCqpiClT. " Original acc: 75.00%, attacked acc: 73.50%, dropped acc: 1.50% Original prompt: Evaluate whether the two provided sentences convey the same meaning by answering 'equivalent' or 'not_equivalent'. Attacked prompt: b"Evaluate whether the two provided sentences convey the same meaning by answering 'equivalent' or 'not_equivalent' SvAp8RlOFn. " Original acc: 75.50%, attacked acc: 72.00%, dropped acc: 3.50% Original prompt: In your capacity as a semantic assessment tool, evaluate the provided sentences and classify their meanings as 'equivalent' or 'not_equivalent'. Attacked prompt: b"In your capacity as a semantic assessment tool, evaluate the provided sentences and classify their meanings as 'equivalent' or 'not_equivalent' Kw6nrs57gH. " Original acc: 75.50%, attacked acc: 73.50%, dropped acc: 2.00% Original prompt: As a sentence similarity evaluator, analyze the provided sentences and indicate if their meanings are 'equivalent' or 'not_equivalent'. Attacked prompt: b"As a sentence similarity evaluator, analyze the provided sentences and indicate if their meanings are 'equivalent' or 'not_equivalent' bhrRSokrfa. " Original acc: 75.50%, attacked acc: 74.00%, dropped acc: 1.50% Original prompt: As a linguistic comparator, review the following pair of sentences and determine their semantic equivalence by choosing 'equivalent' or 'not_equivalent'. Attacked prompt: b"As a linguistic comparator, review the following pair of sentences and determine their semantic equivalence by choosing 'equivalent' or 'not_equivalent' 5XeflW1ZJc. " Original acc: 75.50%, attacked acc: 73.50%, dropped acc: 2.00% ## deepwordbug Original prompt: Can the given sentences be considered semantically identical? Please reply with 'equivalent' or 'not_equivalent'. Attacked prompt: b"an the gimen sentences be considered semantxically identical? Please reply with 'equivalent' or 'not_equivalent'. " Original acc: 76.50%, attacked acc: 72.00%, dropped acc: 4.50% Original prompt: Examine the pair of sentences and decide if their meanings are identical by answering with 'equivalent' or 'not_equivalent'. Attacked prompt: b"Examine the pair of sentences and decide if their maenings are identical by answeroing with 'equivalent' or 'not_equivalent'. " Original acc: 75.50%, attacked acc: 72.00%, dropped acc: 3.50% Original prompt: Evaluate whether the two provided sentences convey the same meaning by answering 'equivalent' or 'not_equivalent'. Attacked prompt: b"Evaluate whether the two provRided sentnces convey the same meaning by answering 'equivalent' or 'not_equivalent'. " Original acc: 74.00%, attacked acc: 70.00%, dropped acc: 4.00% Original prompt: In your capacity as a semantic assessment tool, evaluate the provided sentences and classify their meanings as 'equivalent' or 'not_equivalent'. Attacked prompt: b"nI your capacity as a sematnic asseJssment tol, evaluate the prvoided senRtences and classify their meainngs as 'equivalent' or 'not_equivalent'. " Original acc: 76.00%, attacked acc: 53.00%, dropped acc: 23.00% Original prompt: As a sentence similarity evaluator, analyze the provided sentences and indicate if their meanings are 'equivalent' or 'not_equivalent'. Attacked prompt: b"As a sentence Vsimilarity valuator, analye the proXvided sentences and indicate if their meaniTngs are 'equivalent' or 'not_equivalent'. " Original acc: 75.50%, attacked acc: 60.50%, dropped acc: 15.00% Original prompt: As a linguistic comparator, review the following pair of sentences and determine their semantic equivalence by choosing 'equivalent' or 'not_equivalent'. Attacked prompt: b"As a linguistic compaator, review the following apir of sentences and determine their semantic equvivalence by choosing 'equivalent' or 'not_equivalent'. " Original acc: 75.00%, attacked acc: 72.00%, dropped acc: 3.00% ## stresstest Original prompt: Can the given sentences be considered semantically identical? Please reply with 'equivalent' or 'not_equivalent'. Attacked prompt: b"Can the given sentences be considered semantically identical? Please reply with 'equivalent' or 'not_equivalent' and false is not true . " Original acc: 77.50%, attacked acc: 73.00%, dropped acc: 4.50% Original prompt: Examine the pair of sentences and decide if their meanings are identical by answering with 'equivalent' or 'not_equivalent'. Attacked prompt: b"Examine the pair of sentences and decide if their meanings are identical by answering with 'equivalent' or 'not_equivalent' and true is true and true is true and true is true and true is true and true is true . " Original acc: 75.50%, attacked acc: 74.00%, dropped acc: 1.50% Original prompt: Evaluate whether the two provided sentences convey the same meaning by answering 'equivalent' or 'not_equivalent'. Attacked prompt: b"Evaluate whether the two provided sentences convey the same meaning by answering 'equivalent' or 'not_equivalent' and true is true . " Original acc: 74.50%, attacked acc: 73.00%, dropped acc: 1.50% Original prompt: As a sentence similarity evaluator, analyze the provided sentences and indicate if their meanings are 'equivalent' or 'not_equivalent'. Attacked prompt: b"As a sentence similarity evaluator, analyze the provided sentences and indicate if their meanings are 'equivalent' or 'not_equivalent' and true is true . " Original acc: 75.50%, attacked acc: 74.00%, dropped acc: 1.50% Original prompt: In your capacity as a semantic assessment tool, evaluate the provided sentences and classify their meanings as 'equivalent' or 'not_equivalent'. Attacked prompt: b"In your capacity as a semantic assessment tool, evaluate the provided sentences and classify their meanings as 'equivalent' or 'not_equivalent' and true is true . " Original acc: 76.00%, attacked acc: 15.50%, dropped acc: 60.50% Original prompt: As a linguistic comparator, review the following pair of sentences and determine their semantic equivalence by choosing 'equivalent' or 'not_equivalent'. Attacked prompt: b"As a linguistic comparator, review the following pair of sentences and determine their semantic equivalence by choosing 'equivalent' or 'not_equivalent' and true is true . " Original acc: 75.50%, attacked acc: 74.00%, dropped acc: 1.50% ## textbugger Original prompt: Can the given sentences be considered semantically identical? Please reply with 'equivalent' or 'not_equivalent'. Attacked prompt: b"Ca\xd5\xb8 the given sentences be considered semantically identical? Invites reply with 'equivalent' or 'not_equivalent'. " Original acc: 77.50%, attacked acc: 72.00%, dropped acc: 5.50% Original prompt: Examine the pair of sentences and decide if their meanings are identical by answering with 'equivalent' or 'not_equivalent'. Attacked prompt: b"Examine the pair of s\xd0\xb5ntences and decide if their meani\xd5\xb8gs are same by answering with 'equivalent' or 'not_equivalent'. " Original acc: 75.50%, attacked acc: 70.00%, dropped acc: 5.50% Original prompt: Evaluate whether the two provided sentences convey the same meaning by answering 'equivalent' or 'not_equivalent'. Attacked prompt: b"Evaluate w\xd5\xb0ether the both provided sentences convey the same meanin\xc9\xa1 by answeri\xd5\xb8g 'equivalent' or 'not_equivalent'. " Original acc: 75.00%, attacked acc: 67.50%, dropped acc: 7.50% Original prompt: As a sentence similarity evaluator, analyze the provided sentences and indicate if their meanings are 'equivalent' or 'not_equivalent'. Attacked prompt: b"As a sentencing metaphor evaluator, analyze the provded sentence and depicts if their meanings are 'equivalent' or 'not_equivalent'. " Original acc: 75.50%, attacked acc: 11.50%, dropped acc: 64.00% Original prompt: As a linguistic comparator, review the following pair of sentences and determine their semantic equivalence by choosing 'equivalent' or 'not_equivalent'. Attacked prompt: b"As a linguistic compa\xe2\xb2\x85ator, review the follow\xd1\x96ng p air of sentences and determine their seman\xf0\x9d\x9a\x9dic equivalence by choosing 'equivalent' or 'not_equivalent'. " Original acc: 75.50%, attacked acc: 70.50%, dropped acc: 5.00% Original prompt: In your capacity as a semantic assessment tool, evaluate the provided sentences and classify their meanings as 'equivalent' or 'not_equivalent'. Attacked prompt: b"In your capacities as a sem\xc9\x91ntic assessment tool, evaluate the provdied penalties and classify their meaning as 'equivalent' or 'not_equivalent'. " Original acc: 75.00%, attacked acc: 9.00%, dropped acc: 66.00% ## textfooler Original prompt: Can the given sentences be considered semantically identical? Please reply with 'equivalent' or 'not_equivalent'. Attacked prompt: b"Kan the given conviction be believed semantically identical? Urging reply with 'equivalent' or 'not_equivalent'. " Original acc: 77.00%, attacked acc: 71.50%, dropped acc: 5.50% Original prompt: Examine the pair of sentences and decide if their meanings are identical by answering with 'equivalent' or 'not_equivalent'. Attacked prompt: b"Examine the pair of sentences and selects if their regard are identical by responding with 'equivalent' or 'not_equivalent'. " Original acc: 75.00%, attacked acc: 68.50%, dropped acc: 6.50% Original prompt: Evaluate whether the two provided sentences convey the same meaning by answering 'equivalent' or 'not_equivalent'. Attacked prompt: b"Evaluate whether the two delivered priors convey the same connotation by answering 'equivalent' or 'not_equivalent'. " Original acc: 75.50%, attacked acc: 69.50%, dropped acc: 6.00% Original prompt: In your capacity as a semantic assessment tool, evaluate the provided sentences and classify their meanings as 'equivalent' or 'not_equivalent'. Attacked prompt: b"About your expertise as a semantic appraisal means, evaluate the provided penalty and classify their meanings as 'equivalent' or 'not_equivalent'. " Original acc: 75.50%, attacked acc: 0.00%, dropped acc: 75.50% Original prompt: As a sentence similarity evaluator, analyze the provided sentences and indicate if their meanings are 'equivalent' or 'not_equivalent'. Attacked prompt: b"As a sentence metaphor evaluator, analyze the provided penalties and explain if their meanings are 'equivalent' or 'not_equivalent'. " Original acc: 76.00%, attacked acc: 0.00%, dropped acc: 76.00% Original prompt: As a linguistic comparator, review the following pair of sentences and determine their semantic equivalence by choosing 'equivalent' or 'not_equivalent'. Attacked prompt: b"As a linguistic comparator, review the following mated of judging and discern their semantic equivalence by choosing 'equivalent' or 'not_equivalent'. " Original acc: 75.50%, attacked acc: 71.50%, dropped acc: 4.00% ## translation Language: chinese, acc: 69.00%, prompt: As an expert in semantic comparison, evaluates two given sentences to determine whether they are 'equivalent' or 'not_equivalent'. Language: chinese, acc: 73.00%, prompt: From the point of view of semantic comparison, determine whether this pair of sentences is 'equivalent' or 'not_equivalent'. Language: chinese, acc: 74.00%, prompt: Using the semantic comparison method, determine whether the following two statements are 'equivalent' or 'not_equivalent'. Language: chinese, acc: 75.50%, prompt: For the following two sentences, determine whether they are 'equivalent' or 'not_equivalent' based on semantic comparison. Language: chinese, acc: 62.00%, prompt: As an expert in semantic comparison, please evaluate the following two sentences and determine if they are 'equivalent' or 'not_equivalent'. Language: chinese, acc: 72.50%, prompt: Using semantic comparison techniques, determine whether the following two sentences are 'equivalent' or 'not_equivalent'. Language: chinese, acc: 73.50%, prompt: Please determine whether the following two sentences are 'equivalent' or 'not_equivalent' according to the standard of semantic comparison. Language: chinese, acc: 65.50%, prompt: As an expert in the field of semantic comparison, please evaluate the following two sentences and determine whether they are 'equivalent' or 'not_equivalent'. Language: chinese, acc: 71.00%, prompt: Using semantic comparison, determine whether the following two sentences are 'equivalent' or 'not_equivalent'. Language: chinese, acc: 75.50%, prompt: Determine whether the following two sentences are 'equivalent' or 'not_equivalent' based on semantic comparison. Language: french, acc: 66.00%, prompt: As an expert in semantic comparison, evaluate the following pair of sentences and determine whether they are 'equivalent' or 'not_equivalent'. Language: french, acc: 71.00%, prompt: Can you determine whether the following two sentences are 'equivalent' or 'not_equivalent' as a semantic comparison expert? Language: french, acc: 71.00%, prompt: Using your expertise in semantic comparison, determine whether the following two sentences are 'equivalent' or 'not_equivalent'. Language: french, acc: 50.50%, prompt: As a semantic comparison specialist, assess the similarity between the following two sentences and determine whether they are 'equivalent' or 'not_equivalent'. Language: french, acc: 70.00%, prompt: Are you able to determine whether the following two sentences are 'equivalent' or 'not_equivalent' as an expert in semantic comparison? Language: french, acc: 73.00%, prompt: As a semantic comparison professional, evaluate the following pair of sentences and indicate whether they are 'equivalent' or 'not_equivalent'. Language: french, acc: 73.00%, prompt: Can you determine whether the following two sentences have a 'equivalent' or 'not_equivalent' meaning as an expert in semantic comparison? Language: french, acc: 7.00%, prompt: As an expert in semantic comparison, assess the similarity between the following two sentences and determine whether they are 'equivalent' or 'not_equivalent'. Language: french, acc: 73.50%, prompt: Using your expertise in semantic comparison, determine whether the following two sentences are 'equivalent' or 'not_equivalent' in terms of meaning. Language: french, acc: 59.00%, prompt: As a semantic comparison professional, assess the similarity between the following two sentences and indicate whether they are 'equivalent' or 'not_equivalent'. Language: arabic, acc: 48.50%, prompt: As an expert in semantic comparison, evaluate the two given sentences and determine whether they are 'equivalent' or 'not_equivalent'. Language: arabic, acc: 75.50%, prompt: Based on my experience in semantic analysis, classify the following two sentences as 'equivalent' or 'not_equivalent'. Language: arabic, acc: 67.50%, prompt: As an expert in semantic comparison, analyze the following two sentences and classify them as 'equivalent' or 'not_equivalent'. Language: arabic, acc: 71.50%, prompt: Your task as an expert in semantic comparison is to evaluate the following two sentences and determine whether they are 'equivalent' or 'not_equivalent'. Language: arabic, acc: 75.00%, prompt: As a semantic comparison specialist, analyze the two data statements and insert them into one of the following categories: 'equivalent' or 'not_equivalent'. Language: arabic, acc: 73.50%, prompt: Based on my experience in semantic analysis, classify the following two sentences between 'equivalent' or 'not_equivalent'. Language: arabic, acc: 75.00%, prompt: Your role as a semantic comparison specialist requires analyzing the two given sentences and determining whether they are 'equivalent' or 'not_equivalent'. Language: arabic, acc: 74.00%, prompt: As an experienced semantic analyst, classify the following two sentences as 'equivalent' or 'not_equivalent'. Language: arabic, acc: 73.50%, prompt: Your job as a semantic analyst evaluates the following two sentences as 'equivalent' or 'not_equivalent'. Language: arabic, acc: 75.00%, prompt: As a semantic analyst, determine whether the given sentences are 'equivalent' or 'not_equivalent' based on their relationship. Language: spanish, acc: 77.50%, prompt: As an expert in semantic comparison, it evaluates the pair of sentences provided and determines whether they are 'equivalent' or 'not_equivalent'. Language: spanish, acc: 75.50%, prompt: Based on my experience in semantic analysis, classify the following two sentences as 'equivalent' or 'not_equivalent'. Language: spanish, acc: 59.50%, prompt: As an expert in semantic comparison, analyze the two sentences given and classify them as 'equivalent' or 'not_equivalent'. Language: spanish, acc: 73.50%, prompt: Your task as a semantic comparison specialist is to evaluate the following two sentences and determine whether they are 'equivalent' or 'not_equivalent'. Language: spanish, acc: 69.50%, prompt: As an expert in semantic analysis, he makes a classification of the following two sentences based on their 'equivalent' or 'not_equivalent'. Language: spanish, acc: 76.00%, prompt: Based on your experience of semantic comparison, classify the next two sentences as 'equivalent' or 'not_equivalent'. Language: spanish, acc: 74.50%, prompt: As a specialist in semantic analysis, you are given the task of analysing the two sentences given and classifying them as 'equivalent' or 'not_equivalent'. Language: spanish, acc: 71.50%, prompt: As an expert in semantic comparison, he classifies the following two sentences into 'equivalent' or 'not_equivalent'. Language: spanish, acc: 69.50%, prompt: As a specialist in semantic analysis, evaluate the following two sentences and classify them as 'equivalent' or 'not_equivalent'. Language: spanish, acc: 75.00%, prompt: Your task as an expert in semantic comparison is to analyze the two sentences provided and determine whether they are 'equivalent' or 'not_equivalent' based on their semantic relationship. Language: japanese, acc: 74.00%, prompt: Evaluate whether a given pair of sentences is 'equivalent' or 'not_equivalent', depending on the context. Language: japanese, acc: 78.00%, prompt: Use a semantic comparison to determine whether a given pair of sentences is 'equivalent' or 'not_equivalent'. Language: japanese, acc: 76.50%, prompt: Evaluate a given pair of sentences as 'equivalent' or 'not_equivalent' by determining whether they have the same semantic meaning. Language: japanese, acc: 14.00%, prompt: Determine whether a given pair of sentences is synonyms and evaluate whether they are 'equivalent' or 'not_equivalent'. Language: japanese, acc: 78.00%, prompt: Determine whether a given pair of sentences is 'equivalent' or 'not_equivalent', and whether they are semantically identical. Language: japanese, acc: 35.00%, prompt: Determinate whether a given pair of sentences has the same meaning and evaluate whether they are 'equivalent' or 'not_equivalent'. Language: japanese, acc: 73.00%, prompt: Evaluate whether a given pair of sentences is 'equivalent' or 'not_equivalent' by determining whether they are semantically identical. Language: japanese, acc: 26.50%, prompt: Judge whether a given pair of sentences is equal and evaluate whether they are 'equivalent' or 'not_equivalent'. Language: japanese, acc: 13.00%, prompt: Determinate whether a given pair of sentences are semantically equal and evaluate whether they are 'equivalent' or 'not_equivalent'. Language: japanese, acc: 75.50%, prompt: Whether a given pair of sentences is 'equivalent' or 'not_equivalent' depends on the context. Language: korean, acc: 76.00%, prompt: As a sentence comparator, evaluate the two sentences given to determine 'equivalent' or 'not_equivalent'. Language: korean, acc: 21.00%, prompt: Compare two sentences to determine 'equivalent' or 'not_equivalent'. For this you need qualifications as a specialist in semantic comparison. Language: korean, acc: 68.00%, prompt: It takes your knowledge as an expert in semantic comparison to determine that two sentences are 'equivalent' or 'not_equivalent'. Language: korean, acc: 40.50%, prompt: As a specialist in semantic comparison, evaluate whether two given sentences are 'equivalent' or 'not_equivalent'. Language: korean, acc: 70.50%, prompt: Analyze two sentences to determine 'equivalent' or 'not_equivalent'. For that you need the knowledge of a semantic comparison expert. Language: korean, acc: 79.00%, prompt: As an expert in semantic comparison, decide whether two sentences are 'equivalent' or 'not_equivalent'. Language: korean, acc: 70.00%, prompt: It takes the knowledge of an expert in semantic comparison to compare two sentences to judge 'equivalent' or 'not_equivalent'. Language: korean, acc: 41.50%, prompt: Experience as an expert in semantic comparison is required to determine whether two given sentences are 'equivalent' or 'not_equivalent'. Language: korean, acc: 79.50%, prompt: As an expert in semantic comparison, determine whether two sentences are 'equivalent' or 'not_equivalent'. Language: korean, acc: 12.00%, prompt: Analyze two sentences to determine 'equivalent' or 'not_equivalent'. For this, you need a qualification as a specialist in semantic comparison. # qnli ## 10 prompts Acc: 79.50%, prompt: Assess if the answer to the question can be logically concluded from the provided context by choosing 'entailment' or 'not_entailment'. Acc: 78.50%, prompt: Evaluate whether the given context supports the answer to the question by responding with 'entailment' or 'not_entailment'. Acc: 77.50%, prompt: Consider the context and question, and indicate if the answer can be logically deduced from the context by responding with 'entailment' or 'not_entailment'. Acc: 76.50%, prompt: Please assess if the answer to the question can be derived from the given context by selecting 'entailment' or 'not_entailment'. Acc: 76.00%, prompt: Based on the provided context and question, decide if the information supports the answer by responding with 'entailment' or 'not_entailment'. Acc: 72.50%, prompt: Based on the information in the context, decide if the answer to the question is justified by choosing 'entailment' or 'not_entailment'. Acc: 71.50%, prompt: Examine the context and question, and determine if the context logically implies the answer by selecting 'entailment' or 'not_entailment'. Acc: 68.50%, prompt: Given the question and context provided, determine if the answer can be inferred by choosing 'entailment' or 'not_entailment'. Acc: 65.50%, prompt: Review the given context and question, and decide if the context contains enough information to support the answer by selecting 'entailment' or 'not_entailment'. Acc: 49.00%, prompt: Analyze the context and question, and indicate if the context entails the answer by choosing 'entailment' or 'not_entailment'. Acc: 79.00%, prompt: As a textual inference expert, analyze if the answer to the question can be deduced from the provided context and select 'entailment' or 'not_entailment'. Acc: 78.00%, prompt: In the capacity of a language specialist, decide if the context presented contains enough information to infer the answer to the question and respond with 'entailment' or 'not_entailment'. Acc: 77.00%, prompt: As a semantic researcher, evaluate whether the provided context supports the answer to the question and choose 'entailment' or 'not_entailment'. Acc: 75.50%, prompt: As a linguistic consultant, decide if the answer to the question is logically supported by the provided context and respond with 'entailment' or 'not_entailment'. Acc: 74.50%, prompt: In your role as a semantic evaluator, determine if the provided context justifies the answer to the question and answer with 'entailment' or 'not_entailment'. Acc: 73.00%, prompt: In your role as a linguistic investigator, determine if the context given entails the answer to the question and provide your conclusion with 'entailment' or 'not_entailment'. Acc: 72.50%, prompt: As a language expert, assess if the given context entails the answer to the question and respond with 'entailment' or 'not_entailment'. Acc: 71.50%, prompt: As a semantic interpreter, assess whether the provided context supports the answer to the given question and answer with 'entailment' or 'not_entailment'. Acc: 63.50%, prompt: As a textual analyst, examine if the given context logically implies the answer to the question and indicate your decision with 'entailment' or 'not_entailment'. Acc: 63.00%, prompt: In the capacity of a language evaluator, examine if the given context justifies the answer to the question and indicate your assessment with 'entailment' or 'not_entailment'. ## bertattack Original prompt: Assess if the answer to the question can be logically concluded from the provided context by choosing 'entailment' or 'not_entailment'. Attacked prompt: b"and if the answer to the question can be logically concluded from the provided context by avoiding 'entailment' or 'not_entailment'. " Original acc: 79.00%, attacked acc: 0.00%, dropped acc: 79.00% Original prompt: Evaluate whether the given context supports the answer to the question by responding with 'entailment' or 'not_entailment'. Attacked prompt: b"and whether the response context supports the answer to the question by referring with 'entailment' or 'not_entailment'. " Original acc: 79.00%, attacked acc: 15.00%, dropped acc: 64.00% Original prompt: Consider the context and question, and indicate if the answer can be logically deduced from the context by responding with 'entailment' or 'not_entailment'. Attacked prompt: b"include the context and answered, and signal if the assertion can be logically deduced from the context by following with 'entailment' or 'not_entailment'. " Original acc: 76.50%, attacked acc: 0.00%, dropped acc: 76.50% Original prompt: As a textual inference expert, analyze if the answer to the question can be deduced from the provided context and select 'entailment' or 'not_entailment'. Attacked prompt: b"As a textual inference expert, analysis if the answer to the question can be deduced from the provided citation and define 'entailment' or 'not_entailment'. " Original acc: 79.50%, attacked acc: 0.00%, dropped acc: 79.50% Original prompt: In the capacity of a language specialist, decide if the context presented contains enough information to infer the answer to the question and respond with 'entailment' or 'not_entailment'. Attacked prompt: b"what the capacity of a language student, assess if the context presented contains enough information to infer the answer to the question and conclude with 'entailment' or 'not_entailment'. " Original acc: 78.50%, attacked acc: 0.00%, dropped acc: 78.50% Original prompt: As a semantic researcher, evaluate whether the provided context supports the answer to the question and choose 'entailment' or 'not_entailment'. Attacked prompt: b"As a semantic researcher, show did the provided context supports the answer to the question and include 'entailment' or 'not_entailment'. " Original acc: 77.50%, attacked acc: 0.00%, dropped acc: 77.50% ## checklist Original prompt: Evaluate whether the given context supports the answer to the question by responding with 'entailment' or 'not_entailment'. Attacked prompt: b"Evaluate whether the given context supports the answer to the question by responding with 'entailment' or 'not_entailment' ItGDrrA1Sc. " Original acc: 78.00%, attacked acc: 75.50%, dropped acc: 2.50% Original prompt: Assess if the answer to the question can be logically concluded from the provided context by choosing 'entailment' or 'not_entailment'. Attacked prompt: b"Assess if the answer to the question can be logically concluded from the provided context by choosing 'entailment' or 'not_entailment' 5XeflW1ZJc. " Original acc: 80.50%, attacked acc: 78.50%, dropped acc: 2.00% Original prompt: Consider the context and question, and indicate if the answer can be logically deduced from the context by responding with 'entailment' or 'not_entailment'. Attacked prompt: b"Consider the context and question, and indicate if the answer can be logically deduced from the context by responding with 'entailment' or 'not_entailment' Q47Jmd4lMV. " Original acc: 77.50%, attacked acc: 76.00%, dropped acc: 1.50% Original prompt: As a textual inference expert, analyze if the answer to the question can be deduced from the provided context and select 'entailment' or 'not_entailment'. Attacked prompt: b"As a textual inference expert, analyze if the answer to the question can be deduced from the provided context and select 'entailment' or 'not_entailment' CuVpbbkC4c. " Original acc: 79.00%, attacked acc: 78.00%, dropped acc: 1.00% Original prompt: In the capacity of a language specialist, decide if the context presented contains enough information to infer the answer to the question and respond with 'entailment' or 'not_entailment'. Attacked prompt: b"In the capacity of a language specialist, decide if the context presented contains enough information to infer the answer to the question and respond with 'entailment' or 'not_entailment' OB4KVJzIft. " Original acc: 78.50%, attacked acc: 75.00%, dropped acc: 3.50% Original prompt: As a semantic researcher, evaluate whether the provided context supports the answer to the question and choose 'entailment' or 'not_entailment'. Attacked prompt: b"As a semantic researcher, evaluate whether the provided context supports the answer to the question and choose 'entailment' or 'not_entailment' 6a4Yn3RGVc. " Original acc: 78.00%, attacked acc: 74.50%, dropped acc: 3.50% ## deepwordbug Original prompt: Assess if the answer to the question can be logically concluded from the provided context by choosing 'entailment' or 'not_entailment'. Attacked prompt: b"Asjess if the answer to the question can be logically cocnluded from the provided context by ychoosing 'entailment' or 'not_entailment'. " Original acc: 81.00%, attacked acc: 54.00%, dropped acc: 27.00% Original prompt: Evaluate whether the given context supports the answer to the question by responding with 'entailment' or 'not_entailment'. Attacked prompt: b"Exaluate whetxher the iven contet Supports the anwser to the queston by respEonding with 'entailment' or 'not_entailment'. " Original acc: 78.00%, attacked acc: 50.00%, dropped acc: 28.00% Original prompt: Consider the context and question, and indicate if the answer can be logically deduced from the context by responding with 'entailment' or 'not_entailment'. Attacked prompt: b"ConsideYr the context and Zquestion, and indicaute if the ansUer can be logically deBduced from the congext by qesponding with 'entailment' or 'not_entailment'. " Original acc: 77.50%, attacked acc: 57.50%, dropped acc: 20.00% Original prompt: As a textual inference expert, analyze if the answer to the question can be deduced from the provided context and select 'entailment' or 'not_entailment'. Attacked prompt: b"As a textula inference expert, aInalyze if the answer to the question can be deducud from the provided context and Melect 'entailment' or 'not_entailment'. " Original acc: 79.50%, attacked acc: 19.50%, dropped acc: 60.00% Original prompt: As a semantic researcher, evaluate whether the provided context supports the answer to the question and choose 'entailment' or 'not_entailment'. Attacked prompt: b"As a semantTic reseacrher, Wevaluate whteher the provided context supports the ansewr to the question and dchoose 'entailment' or 'not_entailment'. " Original acc: 77.50%, attacked acc: 49.50%, dropped acc: 28.00% Original prompt: In the capacity of a language specialist, decide if the context presented contains enough information to infer the answer to the question and respond with 'entailment' or 'not_entailment'. Attacked prompt: b"Ii the capacity of a languaFge specPialist, decidY if the cfntext presented contains enoug information to infer the answe to the questio and respond with 'entailment' or 'not_entailment'. " Original acc: 78.00%, attacked acc: 41.00%, dropped acc: 37.00% ## stresstest Original prompt: Assess if the answer to the question can be logically concluded from the provided context by choosing 'entailment' or 'not_entailment'. Attacked prompt: b"Assess if the answer to the question can be logically concluded from the provided context by choosing 'entailment' or 'not_entailment' and true is true . " Original acc: 79.50%, attacked acc: 77.00%, dropped acc: 2.50% Original prompt: Evaluate whether the given context supports the answer to the question by responding with 'entailment' or 'not_entailment'. Attacked prompt: b"Evaluate whether the given context supports the answer to the question by responding with 'entailment' or 'not_entailment' and true is true . " Original acc: 78.50%, attacked acc: 74.50%, dropped acc: 4.00% Original prompt: Consider the context and question, and indicate if the answer can be logically deduced from the context by responding with 'entailment' or 'not_entailment'. Attacked prompt: b"Consider the context and question, and indicate if the answer can be logically deduced from the context by responding with 'entailment' or 'not_entailment' and false is not true . " Original acc: 77.00%, attacked acc: 75.50%, dropped acc: 1.50% Original prompt: As a semantic researcher, evaluate whether the provided context supports the answer to the question and choose 'entailment' or 'not_entailment'. Attacked prompt: b"As a semantic researcher, evaluate whether the provided context supports the answer to the question and choose 'entailment' or 'not_entailment' and true is true and true is true and true is true and true is true and true is true . " Original acc: 77.00%, attacked acc: 72.00%, dropped acc: 5.00% Original prompt: In the capacity of a language specialist, decide if the context presented contains enough information to infer the answer to the question and respond with 'entailment' or 'not_entailment'. Attacked prompt: b"In the capacity of a language specialist, decide if the context presented contains enough information to infer the answer to the question and respond with 'entailment' or 'not_entailment' and true is true . " Original acc: 78.00%, attacked acc: 73.50%, dropped acc: 4.50% Original prompt: As a textual inference expert, analyze if the answer to the question can be deduced from the provided context and select 'entailment' or 'not_entailment'. Attacked prompt: b"As a textual inference expert, analyze if the answer to the question can be deduced from the provided context and select 'entailment' or 'not_entailment' and true is true and true is true and true is true and true is true and true is true . " Original acc: 79.50%, attacked acc: 69.00%, dropped acc: 10.50% ## textbugger Original prompt: Assess if the answer to the question can be logically concluded from the provided context by choosing 'entailment' or 'not_entailment'. Attacked prompt: b"Asse ss if the answer to the question can be logic\xc9\x91lly concluded from the given context by choosing 'entailment' or 'not_entailment'. " Original acc: 81.00%, attacked acc: 60.00%, dropped acc: 21.00% Original prompt: Evaluate whether the given context supports the answer to the question by responding with 'entailment' or 'not_entailment'. Attacked prompt: b"Appraise whether the given context spuports the answer to the question by r\xd0\xb5sponding with 'entailment' or 'not_entailment'. " Original acc: 78.00%, attacked acc: 65.00%, dropped acc: 13.00% Original prompt: Consider the context and question, and indicate if the answer can be logically deduced from the context by responding with 'entailment' or 'not_entailment'. Attacked prompt: b"Consid\xd0\xb5r the context and question, and idicate if the respond can be logically \xd4\x81educed from the context by \xe2\xb2\x85esponding with 'entailment' or 'not_entailment'. " Original acc: 77.00%, attacked acc: 57.00%, dropped acc: 20.00% Original prompt: As a textual inference expert, analyze if the answer to the question can be deduced from the provided context and select 'entailment' or 'not_entailment'. Attacked prompt: b"Because a textual inference expert, analysing if the answe\xe2\xb2\x85 to the question can be deduced from the provid\xd0\xb5d contex\xf0\x9d\x9a\x9d and choosing 'entailment' or 'not_entailment'. " Original acc: 78.50%, attacked acc: 20.00%, dropped acc: 58.50% Original prompt: In the capacity of a language specialist, decide if the context presented contains enough information to infer the answer to the question and respond with 'entailment' or 'not_entailment'. Attacked prompt: b"Onto the capacities of a language specialist, decided if the context presented contain\xd1\x95 enough information to infer the answer to the q\xd5\xbdestion and cater with 'entailment' or 'not_entailment'. " Original acc: 78.00%, attacked acc: 42.50%, dropped acc: 35.50% Original prompt: As a semantic researcher, evaluate whether the provided context supports the answer to the question and choose 'entailment' or 'not_entailment'. Attacked prompt: b"Como a semantic researcher, eval uate whether the provided context supports the answer to the question and chosen 'entailment' or 'not_entailment'. " Original acc: 77.50%, attacked acc: 46.00%, dropped acc: 31.50% ## textfooler Original prompt: Assess if the answer to the question can be logically concluded from the provided context by choosing 'entailment' or 'not_entailment'. Attacked prompt: b"Appraisal if the cater to the question can is obviously finalised from the offer context by opting 'entailment' or 'not_entailment'. " Original acc: 79.00%, attacked acc: 39.00%, dropped acc: 40.00% Original prompt: Evaluate whether the given context supports the answer to the question by responding with 'entailment' or 'not_entailment'. Attacked prompt: b"Evaluating whether the given regard supports the answer to the problema by reactions with 'entailment' or 'not_entailment'. " Original acc: 78.00%, attacked acc: 26.00%, dropped acc: 52.00% Original prompt: Consider the context and question, and indicate if the answer can be logically deduced from the context by responding with 'entailment' or 'not_entailment'. Attacked prompt: b"Recital the context and items, and specify if the solution can be logically subtracted from the framework by reactions with 'entailment' or 'not_entailment'. " Original acc: 77.50%, attacked acc: 0.50%, dropped acc: 77.00% Original prompt: As a textual inference expert, analyze if the answer to the question can be deduced from the provided context and select 'entailment' or 'not_entailment'. Attacked prompt: b"As a textual inference expertise, scans if the satisfy to the item can are subtracted from the conferred background and click 'entailment' or 'not_entailment'. " Original acc: 79.50%, attacked acc: 5.50%, dropped acc: 74.00% Original prompt: As a semantic researcher, evaluate whether the provided context supports the answer to the question and choose 'entailment' or 'not_entailment'. Attacked prompt: b"As a semantic fellows, measuring whether the rendered background supports the answers to the item and chosen 'entailment' or 'not_entailment'. " Original acc: 78.00%, attacked acc: 0.00%, dropped acc: 78.00% Original prompt: In the capacity of a language specialist, decide if the context presented contains enough information to infer the answer to the question and respond with 'entailment' or 'not_entailment'. Attacked prompt: b"Through the potential of a language specialises, opt if the context produced embracing enough info to subtract the answers to the things and cope with 'entailment' or 'not_entailment'. " Original acc: 78.00%, attacked acc: 0.00%, dropped acc: 78.00% ## translation Language: chinese, acc: 71.50%, prompt: Determine whether the answer is implied or not based on the context. Answer 'entailment' or 'not_entailment'. Language: chinese, acc: 78.00%, prompt: As a language expert, determine whether the text contains the answer and choose 'entailment' or 'not_entailment'. Language: chinese, acc: 80.00%, prompt: Determine whether the text implies an answer, and select 'entailment' or 'not_entailment'. Language: chinese, acc: 73.50%, prompt: Given the text and the question, whether the answer is implied, select 'entailment' or 'not_entailment'. Language: chinese, acc: 78.00%, prompt: Determine whether the text contains the answer, depending on the context. Select 'entailment' or 'not_entailment'. Language: chinese, acc: 68.50%, prompt: As a language expert, determine whether a text can contain an answer, and say 'entailment' or 'not_entailment'. Language: chinese, acc: 78.00%, prompt: Please determine whether the text implies an answer. Answer 'entailment' or 'not_entailment'. Language: chinese, acc: 73.50%, prompt: Please select 'entailment' or 'not_entailment' based on the text and the question. Language: chinese, acc: 75.50%, prompt: Assess whether the answer is implied based on the context. Answer 'entailment' or 'not_entailment'. Language: chinese, acc: 76.50%, prompt: Please determine whether the text contains the answer and answer 'entailment' or 'not_entailment'. Language: french, acc: 81.00%, prompt: As a linguistic expert, assess whether the given context involves the answer to the question and answer with 'entailment' or 'not_entailment'. Language: french, acc: 66.00%, prompt: Determine whether the information provided in the context necessarily leads to the answer to the question asked and indicate 'entailment' or 'not_entailment'. Language: french, acc: 57.00%, prompt: Analyze the text to determine if the answer to the question is implied in the context and specify 'entailment' or 'not_entailment'. Language: french, acc: 71.00%, prompt: Based on the given context, decide whether the answer to the question is necessarily involved and mark 'entailment' or 'not_entailment'. Language: french, acc: 74.00%, prompt: Evaluate whether the answer to the question can be deduced from the given context and mark 'entailment' or 'not_entailment'. Language: french, acc: 78.50%, prompt: Discern whether the context provided directly involves the answer to the question and indicate 'entailment' or 'not_entailment'. Language: french, acc: 59.50%, prompt: Determine if the context contains enough information to involve the answer to the question and mark 'entailment' or 'not_entailment'. Language: french, acc: 77.00%, prompt: Assess whether the context provided necessarily leads to the answer to the question and answer with 'entailment' or 'not_entailment'. Language: french, acc: 65.00%, prompt: Analyze the text to determine if the answer to the question is involved in the context and indicate 'entailment' or 'not_entailment'. Language: french, acc: 67.00%, prompt: Based on the given context, decide whether the answer to the question is necessarily inferred and mark 'entailment' or 'not_entailment'. Language: arabic, acc: 67.00%, prompt: As a language expert, evaluate whether the given context calls for an answer and answer 'entailment' or 'not_entailment'. Language: arabic, acc: 69.50%, prompt: Judge the relationship between the text and the question and answer 'entailment' or 'not_entailment', depending on your language experience. Language: arabic, acc: 76.50%, prompt: Does the context given indicate the answer to the question? Evaluate and answer 'entailment' or 'not_entailment'. Language: arabic, acc: 79.50%, prompt: Based on your linguistic knowledge, does the text relate to the question? Answer 'entailment' or 'not_entailment'. Language: arabic, acc: 72.00%, prompt: As a language expert, determine how the text relates to the question and answer 'entailment' or 'not_entailment'. Language: arabic, acc: 74.50%, prompt: Does the text support the answer to the question? Answer 'entailment' or 'not_entailment', depending on your language experience. Language: arabic, acc: 56.50%, prompt: Check the text link to the question and answer 'entailment' or 'not_entailment', depending on your language skills. Language: arabic, acc: 71.50%, prompt: As a language expert, is there a link between the text and the question? Answer 'entailment' or 'not_entailment'. Language: arabic, acc: 61.50%, prompt: Based on your language experience, does context help to answer the question? Evaluate and answer 'entailment' or 'not_entailment'. Language: arabic, acc: 79.50%, prompt: Does the text give a clear answer to the question? Answer 'entailment' or 'not_entailment', depending on your language experience. Language: spanish, acc: 78.50%, prompt: As a language expert, evaluate whether the given context implies the answer to the question and answer with 'entailment' or 'not_entailment'. Language: spanish, acc: 77.00%, prompt: Determine whether the information given in the text necessarily implies the veracity of the hypothesis and answer 'entailment' or 'not_entailment'. Language: spanish, acc: 79.50%, prompt: Analyzes whether the information presented in the paragraph leads to the conclusion of the question and labels the answer as 'entailment' or 'not_entailment'. Language: spanish, acc: 61.00%, prompt: Indicates whether the information provided in the text is sufficient to conclude the statement and labels the response as 'entailment' or 'not_entailment'. Language: spanish, acc: 77.00%, prompt: As an expert on the subject, judge whether the information provided in the text justifies the claim and classify the answer as 'entailment' or 'not_entailment'. Language: spanish, acc: 77.00%, prompt: Evaluates whether the information in the paragraph necessarily supports the conclusion of the hypothesis and responds 'entailment' or 'not_entailment'. Language: spanish, acc: 71.00%, prompt: Determines whether the information presented in the text logically implies the answer to the question and labels the answer as 'entailment' or 'not_entailment'. Language: spanish, acc: 79.50%, prompt: Analyzes whether the information provided in the paragraph necessarily leads to the veracity of the hypothesis and classifies the response as 'entailment' or 'not_entailment'. Language: spanish, acc: 78.00%, prompt: As an expert on the subject, evaluate whether the information presented in the text supports the claim and respond 'entailment' or 'not_entailment'. Language: spanish, acc: 76.00%, prompt: Indicates whether the information provided in the paragraph necessarily implies the answer to the question and labels the answer as 'entailment' or 'not_entailment'. Language: japanese, acc: 76.00%, prompt: Rate whether the answer to the question is derived from the given context and answer with 'entailment' or 'not_entailment'. Language: japanese, acc: 77.00%, prompt: Please answer 'entailment' or 'not_entailment' for the given context and question. Language: japanese, acc: 79.50%, prompt: Decide whether the answer to the question is derived from the given context and answer 'entailment' or 'not_entailment'. Language: japanese, acc: 77.00%, prompt: Compare the question with the given context and give the answer 'entailment' or 'not_entailment'. Language: japanese, acc: 79.00%, prompt: Determinate whether the given context contains the answer to the question and answer with 'entailment' or 'not_entailment'. Language: japanese, acc: 74.00%, prompt: Estimate the answer of the question from the context and give the answer 'entailment' or 'not_entailment'. Language: japanese, acc: 76.50%, prompt: Determinate whether the given context is relevant to the question and answer with 'entailment' or 'not_entailment'. Language: japanese, acc: 77.50%, prompt: Determine whether the given context is relevant to the question and answer with 'entailment' or 'not_entailment'. Language: japanese, acc: 72.00%, prompt: Determinate whether the given context contains the answer to the question and answer 'entailment' or 'not_entailment'. Language: japanese, acc: 73.50%, prompt: Answer with 'entailment' or 'not_entailment', inferring from the given context. Language: korean, acc: 76.00%, prompt: Determine if a given sentence necessarily implies the meaning of another sentence and answer 'entailment' or 'not_entailment'. Language: korean, acc: 72.00%, prompt: By understanding the relations between sentences, judge whether a given sentence necessarily refers to another sentence and answer with 'entailment' or 'not_entailment'. Language: korean, acc: 81.00%, prompt: Evaluate whether a given text necessarily indicates the meaning of another text and respond with 'entailment' or 'not_entailment'. Language: korean, acc: 74.50%, prompt: Understand the relations of a sentence, to determine whether a given sentence necessarily includes other sentences and answer with 'entailment' or 'not_entailment'. Language: korean, acc: 77.50%, prompt: Judge whether a given content necessarily implies the meaning of another content and answer with 'entailment' or 'not_entailment'. Language: korean, acc: 65.00%, prompt: Grasp the relations between sentences, determine if a given sentence necessarily contains the meaning of another sentence and respond with 'entailment' or 'not_entailment'. Language: korean, acc: 75.00%, prompt: Evaluate whether a given text necessarily refers to another text and answer with 'entailment' or 'not_entailment'. Language: korean, acc: 58.50%, prompt: By comparing the meaning of the sentences, to determine if a given sentence necessarily implies another sentence and answer 'entailment' or 'not_entailment'. Language: korean, acc: 71.50%, prompt: Evaluate whether the contents given necessarily refer to other contents and answer with 'entailment' or 'not_entailment'. Language: korean, acc: 77.50%, prompt: By analyzing the relations between sentences, determine if a given sentence does not necessarily include other sentences and answer with 'entailment' or 'not_entailment'. # qqp ## 10 prompts Acc: 84.00%, prompt: Determine if the given pair of statements can be considered the same by responding with 'equivalent' or 'not_equivalent'. Acc: 82.00%, prompt: Assess whether the following statements are identical in meaning by answering 'equivalent' or 'not_equivalent'. Acc: 81.00%, prompt: Can these two statements be considered equal in meaning? Answer with 'equivalent' or 'not_equivalent'. Acc: 81.00%, prompt: Do the following expressions mean the same thing? Provide your answer as 'equivalent' or 'not_equivalent'. Acc: 81.00%, prompt: Analyze if the given set of sentences have the same connotation by answering with 'equivalent' or 'not_equivalent'. Acc: 80.50%, prompt: Are the following two questions equivalent or not? Answer me with "equivalent" or "not_equivalent". Acc: 80.50%, prompt: Examine the following expressions and tell me if they are alike in meaning by using 'equivalent' or 'not_equivalent'. Acc: 80.50%, prompt: Please indicate if the following pair of sentences share the same meaning by responding with 'equivalent' or 'not_equivalent'. Acc: 80.00%, prompt: Are the meanings of these two phrases the same? Reply with 'equivalent' or 'not_equivalent'. Acc: 78.50%, prompt: Evaluate whether these two phrases have identical meanings and respond with 'equivalent' or 'not_equivalent'. Acc: 78.00%, prompt: Do these two sentences convey the same meaning? Indicate with 'equivalent' or 'not_equivalent'. Acc: 82.50%, prompt: As an instrument for question comparison evaluation, consider the questions and determine if their meaning is the same, responding with 'equivalent' for similar questions or 'not_equivalent' for different questions. Acc: 82.00%, prompt: Functioning as a question similarity evaluation tool, analyze the given questions and decide if they share the same meaning, responding with 'equivalent' or 'not_equivalent'. Acc: 82.00%, prompt: While performing question comparison analysis, classify the similarity of the following questions as 'equivalent' for equivalent questions or 'not_equivalent' for different questions. Acc: 82.00%, prompt: Working as a question classifier, identify whether the given questions share the same meaning, and respond with 'equivalent' for equivalent questions or 'not_equivalent' for different ones. Acc: 81.50%, prompt: As a question equivalence detection system, examine the provided questions and respond with 'equivalent' if they are the same in meaning, or 'not_equivalent' if they are different. Acc: 81.50%, prompt: As a tool for determining question equivalence, review the questions and categorize their similarity as either 'equivalent' or 'not_equivalent'. Acc: 81.00%, prompt: In the capacity of a question assessment system, indicate if the meaning of the provided questions is the same, responding with 'equivalent' or 'not_equivalent'. Acc: 81.00%, prompt: In the role of a question similarity analyzer, examine the meaning of the provided questions and decide if they are equivalent, answering with 'equivalent' for equivalent questions or 'not_equivalent' for different questions. Acc: 80.00%, prompt: Acting as a question equivalence instrument, determine if the provided questions are equivalent in meaning, answering with 'equivalent' for similar questions or 'not_equivalent' for dissimilar ones. Acc: 76.50%, prompt: In your role as a question comparison tool, assess the following pair of questions and classify them as 'equivalent' or 'not_equivalent'. ## bertattack Original prompt: Determine if the given pair of statements can be considered the same by responding with 'equivalent' or 'not_equivalent'. Attacked prompt: b"Determine if the any pair of functions can be a the same by responded with 'equivalent' or 'not_equivalent'. " Original acc: 84.00%, attacked acc: 48.50%, dropped acc: 35.50% Original prompt: Assess whether the following statements are identical in meaning by answering 'equivalent' or 'not_equivalent'. Attacked prompt: b"and whether the following statements are difference in meaning by asking 'equivalent' or 'not_equivalent'. " Original acc: 82.50%, attacked acc: 57.00%, dropped acc: 25.50% Original prompt: Can these two statements be considered equal in meaning? Answer with 'equivalent' or 'not_equivalent'. Attacked prompt: b"Can these two constructions be considered equal in meaning? object with 'equivalent' or 'not_equivalent'. " Original acc: 79.50%, attacked acc: 1.00%, dropped acc: 78.50% Original prompt: While performing question comparison analysis, classify the similarity of the following questions as 'equivalent' for equivalent questions or 'not_equivalent' for different questions. Attacked prompt: b"be comparison math comparative analysis, classify the similarities of the some equations as 'equivalent' for equivalent problems or 'not_equivalent' for different questions. " Original acc: 82.00%, attacked acc: 1.50%, dropped acc: 80.50% Original prompt: As an instrument for question comparison evaluation, consider the questions and determine if their meaning is the same, responding with 'equivalent' for similar questions or 'not_equivalent' for different questions. Attacked prompt: b"give an table for category comparison evaluation, consider the questions and determine if their concept is the same, responding with 'equivalent' for similarity questions or 'not_equivalent' for different knowledge. " Original acc: 82.50%, attacked acc: 0.00%, dropped acc: 82.50% Original prompt: Functioning as a question similarity evaluation tool, analyze the given questions and decide if they share the same meaning, responding with 'equivalent' or 'not_equivalent'. Attacked prompt: b"Functioning as a question similarity activation tool, identify the following questions and save if they share the same meanings, responding with 'equivalent' or 'not_equivalent'. " Original acc: 80.50%, attacked acc: 73.50%, dropped acc: 7.00% ## checklist Original prompt: Determine if the given pair of statements can be considered the same by responding with 'equivalent' or 'not_equivalent'. Attacked prompt: b"Determine if the given pair of statements can be considered the same by responding with 'equivalent' or 'not_equivalent' MQnugHcaoy. " Original acc: 83.50%, attacked acc: 80.00%, dropped acc: 3.50% Original prompt: Assess whether the following statements are identical in meaning by answering 'equivalent' or 'not_equivalent'. Attacked prompt: b"Assess whether the following statements are identical in meaning by answering 'equivalent' or 'not_equivalent' iia2yL9Uzm. " Original acc: 81.50%, attacked acc: 78.00%, dropped acc: 3.50% Original prompt: Can these two statements be considered equal in meaning? Answer with 'equivalent' or 'not_equivalent'. Attacked prompt: b"Can these two statements be considered equal in meaning? Answer with 'equivalent' or 'not_equivalent' 5XeflW1ZJc. " Original acc: 80.50%, attacked acc: 80.50%, dropped acc: 0.00% Original prompt: As an instrument for question comparison evaluation, consider the questions and determine if their meaning is the same, responding with 'equivalent' for similar questions or 'not_equivalent' for different questions. Attacked prompt: b"As an instrument for question comparison evaluation, consider the questions and determine if their meaning is the same, responding with 'equivalent' for similar questions or 'not_equivalent' for different questions Cev0s49fbe. " Original acc: 83.00%, attacked acc: 80.50%, dropped acc: 2.50% Original prompt: While performing question comparison analysis, classify the similarity of the following questions as 'equivalent' for equivalent questions or 'not_equivalent' for different questions. Attacked prompt: b"While performing question comparison analysis, classify the similarity of the following questions as 'equivalent' for equivalent questions or 'not_equivalent' for different questions CuVpbbkC4c. " Original acc: 82.50%, attacked acc: 80.50%, dropped acc: 2.00% Original prompt: Functioning as a question similarity evaluation tool, analyze the given questions and decide if they share the same meaning, responding with 'equivalent' or 'not_equivalent'. Attacked prompt: b"Functioning as a question similarity evaluation tool, analyze the given questions and decide if they share the same meaning, responding with 'equivalent' or 'not_equivalent' bhrRSokrfa. " Original acc: 81.00%, attacked acc: 81.00%, dropped acc: 0.00% ## deepwordbug Original prompt: Determine if the given pair of statements can be considered the same by responding with 'equivalent' or 'not_equivalent'. Attacked prompt: b"Determine if the givIn par of statemenfs can be cnosidered the same by responding with 'equivalent' or 'not_equivalent'. " Original acc: 84.00%, attacked acc: 80.00%, dropped acc: 4.00% Original prompt: Assess whether the following statements are identical in meaning by answering 'equivalent' or 'not_equivalent'. Attacked prompt: b"Assess whethdr the followng sLtatements are identocal in meuaning by answerinJ 'equivalent' or 'not_equivalent'. " Original acc: 82.00%, attacked acc: 45.00%, dropped acc: 37.00% Original prompt: Can these two statements be considered equal in meaning? Answer with 'equivalent' or 'not_equivalent'. Attacked prompt: b"Can these two statements be considered equal in meaning? Answer with 'equivalent' or 'not_equivalent'. " Original acc: 80.50%, attacked acc: 80.50%, dropped acc: 1.50% Original prompt: As an instrument for question comparison evaluation, consider the questions and determine if their meaning is the same, responding with 'equivalent' for similar questions or 'not_equivalent' for different questions. Attacked prompt: b"Ao an instrument for question comparison evaluation, consider the questions and determine if their meaning is the same, responding with 'equivalent' for similar questions or 'not_equivalent' for different questions. " Original acc: 82.50%, attacked acc: 79.50%, dropped acc: 3.00% Original prompt: While performing question comparison analysis, classify the similarity of the following questions as 'equivalent' for equivalent questions or 'not_equivalent' for different questions. Attacked prompt: b"Wile performing question comparison analysis, classify the similarity of the following questions as 'equivalent' for equivalent questions or 'not_equivalent' for different questions. " Original acc: 82.50%, attacked acc: 81.50%, dropped acc: 1.00% Original prompt: Functioning as a question similarity evaluation tool, analyze the given questions and decide if they share the same meaning, responding with 'equivalent' or 'not_equivalent'. Attacked prompt: b"Functioning as a question similarity evaluation tool, analyze the given questions and decide if they share the same meaning, responding with 'equivalent' or 'not_equivalent'. " Original acc: 80.50%, attacked acc: 80.50%, dropped acc: 2.00% ## stresstest Original prompt: Determine if the given pair of statements can be considered the same by responding with 'equivalent' or 'not_equivalent'. Attacked prompt: b"Determine if the given pair of statements can be considered the same by responding with 'equivalent' or 'not_equivalent' and false is not true . " Original acc: 84.00%, attacked acc: 79.00%, dropped acc: 5.00% Original prompt: Assess whether the following statements are identical in meaning by answering 'equivalent' or 'not_equivalent'. Attacked prompt: b"Assess whether the following statements are identical in meaning by answering 'equivalent' or 'not_equivalent' and true is true . " Original acc: 82.00%, attacked acc: 77.00%, dropped acc: 5.00% Original prompt: Can these two statements be considered equal in meaning? Answer with 'equivalent' or 'not_equivalent'. Attacked prompt: b"Can these two statements be considered equal in meaning? Answer with 'equivalent' or 'not_equivalent' and false is not true . " Original acc: 80.00%, attacked acc: 72.50%, dropped acc: 7.50% Original prompt: While performing question comparison analysis, classify the similarity of the following questions as 'equivalent' for equivalent questions or 'not_equivalent' for different questions. Attacked prompt: b"While performing question comparison analysis, classify the similarity of the following questions as 'equivalent' for equivalent questions or 'not_equivalent' for different questions and false is not true . " Original acc: 82.00%, attacked acc: 82.50%, dropped acc: -0.50% Original prompt: As an instrument for question comparison evaluation, consider the questions and determine if their meaning is the same, responding with 'equivalent' for similar questions or 'not_equivalent' for different questions. Attacked prompt: b"As an instrument for question comparison evaluation, consider the questions and determine if their meaning is the same, responding with 'equivalent' for similar questions or 'not_equivalent' for different questions and false is not true . " Original acc: 83.00%, attacked acc: 81.00%, dropped acc: 2.00% Original prompt: Functioning as a question similarity evaluation tool, analyze the given questions and decide if they share the same meaning, responding with 'equivalent' or 'not_equivalent'. Attacked prompt: b"Functioning as a question similarity evaluation tool, analyze the given questions and decide if they share the same meaning, responding with 'equivalent' or 'not_equivalent' and true is true and true is true and true is true and true is true and true is true . " Original acc: 81.00%, attacked acc: 80.50%, dropped acc: 0.50% ## textbugger Original prompt: Determine if the given pair of statements can be considered the same by responding with 'equivalent' or 'not_equivalent'. Attacked prompt: b"Ascertain if the given pairs of statemnets can be con\xd1\x95idered the same by responding with 'equivalent' or 'not_equivalent'. " Original acc: 84.00%, attacked acc: 79.00%, dropped acc: 5.00% Original prompt: Assess whether the following statements are identical in meaning by answering 'equivalent' or 'not_equivalent'. Attacked prompt: b"Assess whether the following statements are analogous in maning by answe\xe2\xb2\x85ing 'equivalent' or 'not_equivalent'. " Original acc: 81.50%, attacked acc: 44.50%, dropped acc: 37.00% Original prompt: Can these two statements be considered equal in meaning? Answer with 'equivalent' or 'not_equivalent'. Attacked prompt: b"Can these two sta\xf0\x9d\x9a\x9dements be cons\xd1\x96dered equal in meaning? Answer with 'equivalent' or 'not_equivalent'. " Original acc: 80.50%, attacked acc: 77.00%, dropped acc: 3.50% Original prompt: As an instrument for question comparison evaluation, consider the questions and determine if their meaning is the same, responding with 'equivalent' for similar questions or 'not_equivalent' for different questions. Attacked prompt: b"As an instrument for problem comparison ev\xc9\x91luation, consider the ques\xf0\x9d\x9a\x9dions and determine if their meaning is the same, respond with 'equivalent' for similar questi\xd0\xbens or 'not_equivalent' for different questions. " Original acc: 83.00%, attacked acc: 77.50%, dropped acc: 5.50% Original prompt: While performing question comparison analysis, classify the similarity of the following questions as 'equivalent' for equivalent questions or 'not_equivalent' for different questions. Attacked prompt: b"While performing question compar\xd1\x96son analyses, classify the metaphor of the following question\xd1\x95 as 'equivalent' for equivalent question or 'not_equivalent' for different questions. " Original acc: 82.50%, attacked acc: 77.50%, dropped acc: 5.00% Original prompt: Functioning as a question similarity evaluation tool, analyze the given questions and decide if they share the same meaning, responding with 'equivalent' or 'not_equivalent'. Attacked prompt: b"Functioning as a question similarity evaluation tool, analyze the given questions and decide if they share the same meaning, respondin\xc9\xa1 with 'equivalent' or 'not_equivalent'. " Original acc: 80.50%, attacked acc: 78.50%, dropped acc: 2.00% ## textfooler Original prompt: Determine if the given pair of statements can be considered the same by responding with 'equivalent' or 'not_equivalent'. Attacked prompt: b"Defined if the given peers of depositions can fi dealt the same by reaction with 'equivalent' or 'not_equivalent'. " Original acc: 83.50%, attacked acc: 11.00%, dropped acc: 72.50% Original prompt: Assess whether the following statements are identical in meaning by answering 'equivalent' or 'not_equivalent'. Attacked prompt: b"Estimates whether the following words are identical in meaning by reacts 'equivalent' or 'not_equivalent'. " Original acc: 82.00%, attacked acc: 75.00%, dropped acc: 7.00% Original prompt: Can these two statements be considered equal in meaning? Answer with 'equivalent' or 'not_equivalent'. Attacked prompt: b"Kan these two statement is studied same in sens? Reaction with 'equivalent' or 'not_equivalent'. " Original acc: 80.50%, attacked acc: 0.00%, dropped acc: 80.50% Original prompt: As an instrument for question comparison evaluation, consider the questions and determine if their meaning is the same, responding with 'equivalent' for similar questions or 'not_equivalent' for different questions. Attacked prompt: b"As an instrument for question comparison assess, consider the subjects and establish if their meaning is the same, responding with 'equivalent' for similar doubts or 'not_equivalent' for different questions. " Original acc: 81.50%, attacked acc: 78.50%, dropped acc: 3.00% Original prompt: While performing question comparison analysis, classify the similarity of the following questions as 'equivalent' for equivalent questions or 'not_equivalent' for different questions. Attacked prompt: b"While performing question comparison explored, classify the metaphor of the following questions as 'equivalent' for equivalent questions or 'not_equivalent' for different questions. " Original acc: 82.50%, attacked acc: 78.50%, dropped acc: 4.00% Original prompt: Functioning as a question similarity evaluation tool, analyze the given questions and decide if they share the same meaning, responding with 'equivalent' or 'not_equivalent'. Attacked prompt: b"Functioning as a question similarity evaluation tool, analyse the given questions and decide if they share the same feel, responding with 'equivalent' or 'not_equivalent'. " Original acc: 80.50%, attacked acc: 76.50%, dropped acc: 4.00% ## translation Language: chinese, acc: 3.00%, prompt: As a tool for determining the equivalence of problems, the problems are examined and their similarity is classified as 'equivalent' or 'not_equivalent'. Language: chinese, acc: 79.50%, prompt: Please compare the similarity of two sentences and put them into the category of 'equivalent' or 'not_equivalent'. Language: chinese, acc: 76.00%, prompt: Two sentences are classified as 'equivalent' or 'not_equivalent' for their similarity. Language: chinese, acc: 74.50%, prompt: You can determine how similar the questions are by comparing them and categorizing them as 'equivalent' or 'not_equivalent'. Language: chinese, acc: 15.50%, prompt: Using the method of contrast, the similarity of these problems is divided into two categories: 'equivalent' or 'not_equivalent'. Language: chinese, acc: 77.00%, prompt: By comparing these issues, you can classify them as 'equivalent' or 'not_equivalent'. Language: chinese, acc: 81.50%, prompt: To determine whether the questions are similar, put them into the category of 'equivalent' or 'not_equivalent'. Language: chinese, acc: 84.00%, prompt: Divide the similarity of these questions into 'equivalent' or 'not_equivalent' categories. Language: chinese, acc: 81.00%, prompt: Using the similarity assessment tool, these questions were classified as 'equivalent' or 'not_equivalent'. Language: chinese, acc: 75.00%, prompt: By analyzing the similarity of these problems, they are divided into categories of 'equivalent' or 'not_equivalent'. Language: french, acc: 80.00%, prompt: As a tool to determine the equivalence of questions, review the questions and rank their similarity as 'equivalent' or 'not_equivalent'. Language: french, acc: 42.00%, prompt: Please compare the similarity of two sentences and classify them as 'equivalent' or 'not_equivalent'. Language: french, acc: 80.00%, prompt: Based on the similarity of two sentences, classify them as 'equivalent' or 'not_equivalent'. Language: french, acc: 85.00%, prompt: You can determine the similarity between these questions by comparing them and classifying them as 'equivalent' or 'not_equivalent'. Language: french, acc: 66.00%, prompt: Use a comparative method to divide the similarity of these questions into two categories: 'equivalent' or 'not_equivalent'. Language: french, acc: 79.00%, prompt: By comparing these questions, you can classify them as 'equivalent' or 'not_equivalent'. Language: french, acc: 25.00%, prompt: Determine whether these questions are similar or not, and then classify them as 'equivalent' or 'not_equivalent'. Language: french, acc: 82.50%, prompt: Divide the similarity of these questions into two categories: 'equivalent' or 'not_equivalent'. Language: french, acc: 83.00%, prompt: Use a similarity assessment tool to classify these questions as 'equivalent' or 'not_equivalent'. Language: french, acc: 82.50%, prompt: By analyzing the similarity of these questions, you can divide them into two categories: 'equivalent' or 'not_equivalent'. Language: arabic, acc: 55.50%, prompt: As a tool for determining an equation of questions, review the questions and classify their similarity as either 'equivalent' or 'not_equivalent'. Language: arabic, acc: 82.50%, prompt: When using questions in the classification domain, please classify the similarity between the questions as 'equivalent' or 'not_equivalent'. Language: arabic, acc: 6.00%, prompt: To determine an equation of questions, you must review the questions and classify their similarity as 'equivalent' or 'not_equivalent'. Language: arabic, acc: 74.50%, prompt: Questions can be classified as 'equivalent' or 'not_equivalent' when used to identify classifications. Language: arabic, acc: 70.50%, prompt: Classification of question similarity as 'equivalent' or 'not_equivalent' is used as a tool to determine the classification of questions. Language: arabic, acc: 83.00%, prompt: Classify the similarity of the questions as 'equivalent' or 'not_equivalent' to determine the equation of the questions. Language: arabic, acc: 56.50%, prompt: Identifying the similarity of questions and classifying them as 'equivalent' or 'not_equivalent' is an important tool in determining the classification of questions. Language: arabic, acc: 55.00%, prompt: When classifying questions, their similarity can be classified as 'equivalent' or 'not_equivalent' to determine the correct classification. Language: arabic, acc: 83.50%, prompt: The similarity of questions should be classified as 'equivalent' or 'not_equivalent' when used to determine the equation of questions. Language: arabic, acc: 74.50%, prompt: Identifying the similarity of questions and classifying them as 'equivalent' or 'not_equivalent' helps to correctly classify questions. Language: spanish, acc: 23.50%, prompt: As a tool to determine the equivalence of questions, it reviews the questions and classifies their similarity as 'equivalent' or 'not_equivalent'. Language: spanish, acc: 80.00%, prompt: Evaluate the similarity between questions and classify them as 'equivalent' or 'not_equivalent' to determine their equivalence. Language: spanish, acc: 81.50%, prompt: Determine whether two questions are 'equivalent' or 'not_equivalent' based on similarity and characteristics. Language: spanish, acc: 85.00%, prompt: Classifies the similarity between questions as 'equivalent' or 'not_equivalent' to determine their equivalence. Language: spanish, acc: 81.50%, prompt: Review the questions and rate them as 'equivalent' or 'not_equivalent' based on their similarity and content. Language: spanish, acc: 40.50%, prompt: As part of the classification task of questions, it determines their equivalence by categorizing their similarity as 'equivalent' or 'not_equivalent'. Language: spanish, acc: 78.00%, prompt: Analyze the similarity between questions and classify them as 'equivalent' or 'not_equivalent' to determine their equivalence. Language: spanish, acc: 40.00%, prompt: As a method of identifying the equivalence of questions, it categorizes their similarity as 'equivalent' or 'not_equivalent'. Language: spanish, acc: 64.00%, prompt: To determine the equivalence between questions, check their similarity and classify them as 'equivalent' or 'not_equivalent'. Language: spanish, acc: 82.50%, prompt: Classify the similarity between questions as 'equivalent' or 'not_equivalent' to determine whether they are equivalent or not. Language: japanese, acc: 74.50%, prompt: As a tool to determine the equivalence of the question, review the question and categorize its similarities into 'equivalent' or 'not_equivalent' categories. Language: japanese, acc: 52.00%, prompt: Work on text sorting tasks labeled 'equivalent' or 'not_equivalent'. Language: japanese, acc: 50.00%, prompt: For text classification tasks, use the labels 'equivalent' or 'not_equivalent' to determine the equivalence of statements. Language: japanese, acc: 48.50%, prompt: In the MRPC dataset, use the labels 'equivalent' or 'not_equivalent' to classify the equivalence of statements. Language: japanese, acc: 63.50%, prompt: As a tool for determining equivalence, check sentences and categorize them into 'equivalent' or 'not_equivalent' categories. Language: japanese, acc: 67.00%, prompt: Use the labels 'equivalent' or 'not_equivalent' to determine the equivalence of statements in text classification tasks. Language: japanese, acc: 19.50%, prompt: In the text classification task of the MRPC data set, classify the equivalence of statements with labels of 'equivalent' or 'not_equivalent'. Language: japanese, acc: 44.00%, prompt: As a tool to determine the equivalence of statements, categorize statements into 'equivalent' or 'not_equivalent' categories. Language: japanese, acc: 73.50%, prompt: In a text classification task, classify the equivalence of statements using labels of 'equivalent' or 'not_equivalent'. Language: japanese, acc: 77.50%, prompt: Do a text classification task to determine the equivalence of statements, labeled 'equivalent' or 'not_equivalent'. Language: korean, acc: 79.00%, prompt: Classify two given sentences as 'equivalent' or 'not_equivalent' by discriminating whether they have the same meaning. Language: korean, acc: 83.50%, prompt: Determine sentence equivalence by judging the similarity of two sentences with 'equivalent' or 'not_equivalent'. Language: korean, acc: 81.50%, prompt: Classify the similarity of sentences as 'equivalent' or 'not_equivalent' by judging whether two sentences have the same meaning. Language: korean, acc: 47.00%, prompt: Determine if two given sentences are equivalent to each other, and classify their similarity as 'equivalent' or 'not_equivalent'. Language: korean, acc: 69.00%, prompt: Compare two given sentences to determine sentence equivalence, and classify their similarities as 'equivalent' or 'not_equivalent'. Language: korean, acc: 82.00%, prompt: Classify sentence equivalence as 'equivalent' or 'not_equivalent' by judging whether two sentences have the same meaning to each other. Language: korean, acc: 60.50%, prompt: Determine if two sentences have the same meaning, and classify their similarities as 'equivalent' or 'not_equivalent'. Language: korean, acc: 46.50%, prompt: Compare two given sentences to determine their equivalence, and classify their similarities as 'equivalent' or 'not_equivalent'. Language: korean, acc: 24.00%, prompt: Review two sentences to evaluate sentence equivalence, and classify their similarities as 'equivalent' or 'not_equivalent'. Language: korean, acc: 83.00%, prompt: Judge whether two sentences have the same meaning to each other, and determine the sentence equivalence with 'equivalent' or 'not_equivalent'. # rte ## 10 prompts Acc: 84.00%, prompt: Determine if the given pair of sentences displays entailment or not_entailment. Respond with 'entailment' or 'not_entailment'. Acc: 83.00%, prompt: Are the following two sentences entailment or not_entailment? Answer me with "entailment" or "not_entailment", just one word. Acc: 83.00%, prompt: Identify whether the given pair of sentences demonstrates entailment or not_entailment. Answer with 'entailment' or 'not_entailment'. Acc: 81.50%, prompt: Examine the pair of sentences and determine if they exhibit entailment or not_entailment. Answer with either 'entailment' or 'not_entailment'. Acc: 80.50%, prompt: Does the relationship between the given sentences represent entailment or not_entailment? Respond with 'entailment' or 'not_entailment'. Acc: 79.50%, prompt: Please classify the relationship between the provided sentences as 'entailment' or 'not_entailment'. Acc: 78.50%, prompt: Review the two statements and categorize their relationship as either 'entailment' or 'not_entailment'. Acc: 76.50%, prompt: Indicate if the connection between the following sentences is 'entailment' or 'not_entailment'. Acc: 76.00%, prompt: Analyze the two provided sentences and decide if their relationship is 'entailment' or 'not_entailment'. Acc: 71.50%, prompt: Considering the two sentences, identify if their relationship is 'entailment' or 'not_entailment'. Acc: 70.50%, prompt: Assess the connection between the following sentences and classify it as 'entailment' or 'not_entailment'. Acc: 83.00%, prompt: Acting as an entailment detection instrument, determine if the given pair of sentences demonstrates entailment or not_entailment. Answer with 'entailment' or 'not_entailment': Acc: 81.50%, prompt: As an instrument for entailment evaluation, consider the two sentences and determine if their relationship is 'entailment' or 'not_entailment'. Respond with 'entailment' or 'not_entailment': Acc: 81.00%, prompt: Working as an entailment classifier, identify whether the given pair of sentences displays entailment or not_entailment. Respond with 'entailment' or 'not_entailment': Acc: 80.50%, prompt: In the role of a semantic relationship analyzer, examine the connection between the given sentences and decide if they exhibit entailment or not_entailment. Answer with 'entailment' or 'not_entailment': Acc: 79.00%, prompt: While performing entailment analysis, classify the relationship between the provided sentences as 'entailment' or 'not_entailment': Acc: 78.00%, prompt: In the capacity of an entailment assessment system, indicate if the link between the following sentences is 'entailment' or 'not_entailment': Acc: 77.00%, prompt: In your role as an entailment analysis tool, assess the relationship between the given sentences and classify it as 'entailment' or 'not_entailment': Acc: 77.00%, prompt: As a tool for determining entailment relationships, review the two statements and categorize their connection as either 'entailment' or 'not_entailment': Acc: 76.50%, prompt: Functioning as an entailment evaluation tool, analyze the provided sentences and decide if their relationship is 'entailment' or 'not_entailment': Acc: 76.00%, prompt: As an entailment identification system, examine the connection between the following sentences and respond with 'entailment' or 'not_entailment': ## bertattack Original prompt: Are the following two sentences entailment or not_entailment? Answer me with "entailment" or "not_entailment", just one word. Attacked prompt: b'like the following both words entailment or not_entailment? question me with "entailment" or "not_entailment", just one word. ' Original acc: 84.00%, attacked acc: 69.00%, dropped acc: 15.00% Original prompt: Identify whether the given pair of sentences demonstrates entailment or not_entailment. Answer with 'entailment' or 'not_entailment'. Attacked prompt: b"determine if the given pair of sentences denotes entailment or not_entailment. contrast with 'entailment' or 'not_entailment'." Original acc: 83.50%, attacked acc: 76.00%, dropped acc: 7.50% Original prompt: Determine if the given pair of sentences displays entailment or not_entailment. Respond with 'entailment' or 'not_entailment'. Attacked prompt: b"determine if the following singular of sentence indicating entailment or not_entailment. contrast with 'entailment' or 'not_entailment'." Original acc: 82.50%, attacked acc: 55.50%, dropped acc: 27.00% Original prompt: Acting as an entailment detection instrument, determine if the given pair of sentences demonstrates entailment or not_entailment. Answer with 'entailment' or 'not_entailment': Attacked prompt: b"before as an entailment detection instrument, determine if the given person of verbs demonstrating entailment or not_entailment. are with 'entailment' or 'not_entailment':" Original acc: 83.50%, attacked acc: 44.00%, dropped acc: 39.50% Original prompt: Working as an entailment classifier, identify whether the given pair of sentences displays entailment or not_entailment. Respond with 'entailment' or 'not_entailment': Attacked prompt: b"at as an entailment classifier, know which the given type of verb displays entailment or not_entailment. object with 'entailment' or 'not_entailment':" Original acc: 81.50%, attacked acc: 37.50%, dropped acc: 44.00% Original prompt: As an instrument for entailment evaluation, consider the two sentences and determine if their relationship is 'entailment' or 'not_entailment'. Respond with 'entailment' or 'not_entailment': Attacked prompt: b"it an instrument for entailment activation, consider the into sentences and determine if their relationship is 'entailment' or 'not_entailment'. appear with 'entailment' or 'not_entailment':" Original acc: 81.00%, attacked acc: 70.50%, dropped acc: 10.50% ## checklist Original prompt: Are the following two sentences entailment or not_entailment? Answer me with "entailment" or "not_entailment", just one word. Attacked prompt: b'Are the following two sentences entailment or not_entailment? Answer me with "entailment" or "not_entailment", just one word 5XeflW1ZJc. ' Original acc: 83.00%, attacked acc: 80.50%, dropped acc: 2.50% Original prompt: Identify whether the given pair of sentences demonstrates entailment or not_entailment. Answer with 'entailment' or 'not_entailment'. Attacked prompt: b"Identify whether the given pair of sentences demonstrates entailment or not_entailment. Answer with 'entailment' or 'not_entailment' lf8wBa2yEm." Original acc: 83.50%, attacked acc: 80.50%, dropped acc: 3.00% Original prompt: Determine if the given pair of sentences displays entailment or not_entailment. Respond with 'entailment' or 'not_entailment'. Attacked prompt: b"Determine if the given pair of sentences displays entailment or not_entailment. Respond with 'entailment' or 'not_entailment' MSAddJ4D2a." Original acc: 82.50%, attacked acc: 80.00%, dropped acc: 2.50% Original prompt: Acting as an entailment detection instrument, determine if the given pair of sentences demonstrates entailment or not_entailment. Answer with 'entailment' or 'not_entailment': Attacked prompt: b"Acting as an entailment detection instrument, determine if the given pair of sentences demonstrates entailment or not_entailment. Answer with 'entailment' or 'not_entailment' Kw6nrs57gH:" Original acc: 82.50%, attacked acc: 82.00%, dropped acc: 0.50% Original prompt: As an instrument for entailment evaluation, consider the two sentences and determine if their relationship is 'entailment' or 'not_entailment'. Respond with 'entailment' or 'not_entailment': Attacked prompt: b"As an instrument for entailment evaluation, consider the two sentences and determine if their relationship is 'entailment' or 'not_entailment'. Respond with 'entailment' or 'not_entailment' 6a4Yn3RGVc:" Original acc: 81.50%, attacked acc: 80.00%, dropped acc: 1.50% Original prompt: Working as an entailment classifier, identify whether the given pair of sentences displays entailment or not_entailment. Respond with 'entailment' or 'not_entailment': Attacked prompt: b"Working as an entailment classifier, identify whether the given pair of sentences displays entailment or not_entailment. Respond with 'entailment' or 'not_entailment' xikCjkMydH:" Original acc: 81.50%, attacked acc: 80.50%, dropped acc: 1.00% ## deepwordbug Original prompt: Identify whether the given pair of sentences demonstrates entailment or not_entailment. Answer with 'entailment' or 'not_entailment'. Attacked prompt: b"Identify whether the given pair of sentences demonstraets entailment or not_entailment. Answeg with 'entailment' or 'not_entailment'." Original acc: 83.50%, attacked acc: 81.00%, dropped acc: 2.50% Original prompt: Are the following two sentences entailment or not_entailment? Answer me with "entailment" or "not_entailment", just one word. Attacked prompt: b'ArYe the folowing two sentenes entailment or not_entailment? Answer me with "entailment" or "not_entailment", just one word. ' Original acc: 83.50%, attacked acc: 79.00%, dropped acc: 4.50% Original prompt: Determine if the given pair of sentences displays entailment or not_entailment. Respond with 'entailment' or 'not_entailment'. Attacked prompt: b"Determine if the given pair of Wentences displays entailment or not_entailment. Respnod with 'entailment' or 'not_entailment'." Original acc: 82.50%, attacked acc: 80.00%, dropped acc: 2.50% Original prompt: Acting as an entailment detection instrument, determine if the given pair of sentences demonstrates entailment or not_entailment. Answer with 'entailment' or 'not_entailment': Attacked prompt: b"Acting as an entailment detection instrument, determine if the given pair of sentenceys demonstrates entailment or not_entailment. Answer with 'entailment' or 'not_entailment':" Original acc: 83.00%, attacked acc: 80.00%, dropped acc: 3.00% Original prompt: Working as an entailment classifier, identify whether the given pair of sentences displays entailment or not_entailment. Respond with 'entailment' or 'not_entailment': Attacked prompt: b"Working as an entailment classifier, identify whether the given pair of sentences displays entailment or not_entailment. Respond with 'entailment' or 'not_entailment':" Original acc: 80.50%, attacked acc: 80.50%, dropped acc: 2.50% Original prompt: As an instrument for entailment evaluation, consider the two sentences and determine if their relationship is 'entailment' or 'not_entailment'. Respond with 'entailment' or 'not_entailment': Attacked prompt: b"IAs an instrument for entailment evaluatioRn, consider the otwo sentences and determine if their relationship is 'entailment' or 'not_entailment'. Respond with 'entailment' or 'not_entailment':" Original acc: 80.50%, attacked acc: 78.50%, dropped acc: 2.00% ## stresstest Original prompt: Identify whether the given pair of sentences demonstrates entailment or not_entailment. Answer with 'entailment' or 'not_entailment'. Attacked prompt: b"Identify whether the given pair of sentences demonstrates entailment or not_entailment. Answer with 'entailment' or 'not_entailment' and false is not true ." Original acc: 83.50%, attacked acc: 82.50%, dropped acc: 1.00% Original prompt: Are the following two sentences entailment or not_entailment? Answer me with "entailment" or "not_entailment", just one word. Attacked prompt: b'Are the following two sentences entailment or not_entailment? Answer me with "entailment" or "not_entailment", just one word and true is true . ' Original acc: 83.50%, attacked acc: 81.50%, dropped acc: 2.00% Original prompt: Determine if the given pair of sentences displays entailment or not_entailment. Respond with 'entailment' or 'not_entailment'. Attacked prompt: b"Determine if the given pair of sentences displays entailment or not_entailment. Respond with 'entailment' or 'not_entailment' and true is true and true is true and true is true and true is true and true is true ." Original acc: 82.50%, attacked acc: 81.50%, dropped acc: 1.00% Original prompt: Acting as an entailment detection instrument, determine if the given pair of sentences demonstrates entailment or not_entailment. Answer with 'entailment' or 'not_entailment': Attacked prompt: b"Acting as an entailment detection instrument, determine if the given pair of sentences demonstrates entailment or not_entailment. Answer with 'entailment' or 'not_entailment' and true is true :" Original acc: 82.50%, attacked acc: 83.50%, dropped acc: -1.00% Original prompt: Working as an entailment classifier, identify whether the given pair of sentences displays entailment or not_entailment. Respond with 'entailment' or 'not_entailment': Attacked prompt: b"Working as an entailment classifier, identify whether the given pair of sentences displays entailment or not_entailment. Respond with 'entailment' or 'not_entailment' and true is true and true is true and true is true and true is true and true is true :" Original acc: 81.00%, attacked acc: 81.00%, dropped acc: 0.00% Original prompt: As an instrument for entailment evaluation, consider the two sentences and determine if their relationship is 'entailment' or 'not_entailment'. Respond with 'entailment' or 'not_entailment': Attacked prompt: b"As an instrument for entailment evaluation, consider the two sentences and determine if their relationship is 'entailment' or 'not_entailment'. Respond with 'entailment' or 'not_entailment' and true is true and true is true and true is true and true is true and true is true :" Original acc: 81.50%, attacked acc: 80.50%, dropped acc: 1.00% ## textbugger Original prompt: Identify whether the given pair of sentences demonstrates entailment or not_entailment. Answer with 'entailment' or 'not_entailment'. Attacked prompt: b"Identifies whether the given pair of sentences demonstrates entailment or not_entailment. Answer with 'entailment' or 'not_entailment'." Original acc: 83.50%, attacked acc: 80.00%, dropped acc: 3.50% Original prompt: Are the following two sentences entailment or not_entailment? Answer me with "entailment" or "not_entailment", just one word. Attacked prompt: b'Be the following two sentences entailment or not_entailment? Answer me with "entailment" or "not_entailment", just one word. ' Original acc: 83.00%, attacked acc: 78.50%, dropped acc: 4.50% Original prompt: Determine if the given pair of sentences displays entailment or not_entailment. Respond with 'entailment' or 'not_entailment'. Attacked prompt: b"Ascertain if the given pair of sentences displays entailment or not_entailment. Answering with 'entailment' or 'not_entailment'." Original acc: 83.00%, attacked acc: 79.50%, dropped acc: 3.50% Original prompt: Acting as an entailment detection instrument, determine if the given pair of sentences demonstrates entailment or not_entailment. Answer with 'entailment' or 'not_entailment': Attacked prompt: b"Caretaker as an entailment detects instrument, determine if the given pair of punishments illustrates entailment or not_entailment. Responding with 'entailment' or 'not_entailment':" Original acc: 82.50%, attacked acc: 77.50%, dropped acc: 5.00% Original prompt: Working as an entailment classifier, identify whether the given pair of sentences displays entailment or not_entailment. Respond with 'entailment' or 'not_entailment': Attacked prompt: b"Working as an entailment classifier, detecting if the given torque of sentences displays entailment or not_entailment. Answering with 'entailment' or 'not_entailment':" Original acc: 81.00%, attacked acc: 74.50%, dropped acc: 6.50% Original prompt: As an instrument for entailment evaluation, consider the two sentences and determine if their relationship is 'entailment' or 'not_entailment'. Respond with 'entailment' or 'not_entailment': Attacked prompt: b"As an instrumentation for entailment evaluation, consider the two sentences and determine if their nexus is 'entailment' or 'not_entailment'. Respond with 'entailment' or 'not_entailment':" Original acc: 81.00%, attacked acc: 77.00%, dropped acc: 4.00% ## textfooler Original prompt: Are the following two sentences entailment or not_entailment? Answer me with "entailment" or "not_entailment", just one word. Attacked prompt: b'Pose the aftermath two phrases entailment or not_entailment? Reactions me with "entailment" or "not_entailment", just one adjectives. ' Original acc: 83.50%, attacked acc: 60.50%, dropped acc: 23.00% Original prompt: Identify whether the given pair of sentences demonstrates entailment or not_entailment. Answer with 'entailment' or 'not_entailment'. Attacked prompt: b"Detected whether the given pair of sentences demonstrates entailment or not_entailment. Response with 'entailment' or 'not_entailment'." Original acc: 84.00%, attacked acc: 79.50%, dropped acc: 4.50% Original prompt: Determine if the given pair of sentences displays entailment or not_entailment. Respond with 'entailment' or 'not_entailment'. Attacked prompt: b"Detects if the bestowed peers of sentences denotes entailment or not_entailment. Cope with 'entailment' or 'not_entailment'." Original acc: 83.00%, attacked acc: 73.00%, dropped acc: 10.00% Original prompt: Acting as an entailment detection instrument, determine if the given pair of sentences demonstrates entailment or not_entailment. Answer with 'entailment' or 'not_entailment': Attacked prompt: b"Caretaker as an entailment sensing tools, defined if the given peers of convicting depicting entailment or not_entailment. Solution with 'entailment' or 'not_entailment':" Original acc: 83.00%, attacked acc: 68.00%, dropped acc: 15.00% Original prompt: Working as an entailment classifier, identify whether the given pair of sentences displays entailment or not_entailment. Respond with 'entailment' or 'not_entailment': Attacked prompt: b"Acted as an entailment classifier, detecting whether the given peers of chastisement show entailment or not_entailment. Reactions with 'entailment' or 'not_entailment':" Original acc: 81.00%, attacked acc: 65.00%, dropped acc: 16.00% Original prompt: As an instrument for entailment evaluation, consider the two sentences and determine if their relationship is 'entailment' or 'not_entailment'. Respond with 'entailment' or 'not_entailment': Attacked prompt: b"Because an devices for entailment appraised, revisit the two phrases and identifying if their correlate is 'entailment' or 'not_entailment'. Answered with 'entailment' or 'not_entailment':" Original acc: 81.00%, attacked acc: 60.00%, dropped acc: 21.00% ## translation Language: chinese, acc: 76.00%, prompt: In the light of an implication analysis tool, evaluate the relationship between the following sentences and classify them as 'entailment' or 'not_entailment'. Language: chinese, acc: 71.50%, prompt: From the perspective of an implication analysis tool, determine whether there is an implication relationship in the following sentences by classifying them as 'entailment' or 'not_entailment'. Language: chinese, acc: 60.50%, prompt: Please use an implication analysis tool to determine whether an implication relationship exists in the following sentences by classifying them as 'entailment' or 'not_entailment'. Language: chinese, acc: 78.50%, prompt: Please evaluate the relation of the following sentences as 'entailment' or 'not_entailment' for the purpose of determining implication relation. Language: chinese, acc: 70.00%, prompt: Please use the implication analysis tool to evaluate the relationships between the following sentences and classify them as 'entailment' or 'not_entailment'. Language: chinese, acc: 68.50%, prompt: For the purpose of determining implicative relations, analyze the relations of the following sentences and classify them as 'entailment' or 'not_entailment'. Language: chinese, acc: 69.50%, prompt: Please use the implication analysis tool to determine the relationship of the following sentences and classify them as 'entailment' or 'not_entailment'. Language: chinese, acc: 62.50%, prompt: Please use the implication judgment tool to assess the relevance of the following sentences and classify them as 'entailment' or 'not_entailment'. Language: chinese, acc: 73.50%, prompt: Please, with implication analysis as the main task, determine the relationships between the following sentences and classify them as 'entailment' or 'not_entailment'. Language: chinese, acc: 65.50%, prompt: Using the implication judgment as a criterion, analyze the relation of the following sentences and classify them as 'entailment' or 'not_entailment'. Language: french, acc: 80.50%, prompt: As an engagement analysis tool, evaluate the relationship between the given sentences and classify it as 'entailment' or 'not_entailment'. Language: french, acc: 77.50%, prompt: Determine whether the given sentences involve one another or not as an implication analysis tool. Classify them accordingly as 'entailment' or 'not_entailment'. Language: french, acc: 79.50%, prompt: Using implication analysis, evaluate whether the sentences provided have a logical relationship and categorize them as 'entailment' or 'not_entailment'. Language: french, acc: 80.00%, prompt: As an engagement assessment tool, determine whether the sentences provided have a logical relationship and classify them as 'entailment' or 'not_entailment'. Language: french, acc: 73.50%, prompt: As an implication classification tool, analyze the sentences provided to determine if there is a logical relationship and categorize them as 'entailment' or 'not_entailment'. Language: french, acc: 45.00%, prompt: Using implication analysis, determine whether the given sentences have a cause-effect relationship and categorize them as 'entailment' or 'not_entailment'. Language: french, acc: 73.00%, prompt: Evaluate the relationship between the given sentences using implication analysis and rank them accordingly as 'entailment' or 'not_entailment'. Language: french, acc: 74.50%, prompt: As an engagement detection tool, determine whether the given sentences have a logical relationship and categorize them as 'entailment' or 'not_entailment'. Language: french, acc: 61.50%, prompt: Using implication analysis, evaluate whether the sentences provided have a cause-effect relationship and rank them accordingly as 'entailment' or 'not_entailment'. Language: french, acc: 39.00%, prompt: Determine whether the given sentences have a cause-effect relationship as an engagement analysis tool and categorize them as 'entailment' or 'not_entailment'. Language: arabic, acc: 78.50%, prompt: In your role as a tool for reasoning analysis, evaluate the relationship between given sentences and classify them as 'entailment' or 'not_entailment'. Language: arabic, acc: 76.50%, prompt: Can you determine whether this sentence is inferred from the other sentence? Classify it as 'entailment' or 'not_entailment'. Language: arabic, acc: 78.50%, prompt: Using the tool of reasoning analysis, analyze the relationship between given sentences and classify them as 'entailment' or 'not_entailment'. Language: arabic, acc: 69.50%, prompt: Does this sentence represent a conclusion from the previous sentence? Classify it as 'entailment' or 'not_entailment'. Language: arabic, acc: 80.50%, prompt: As a tool of reasoning analysis, evaluate the relationship of given sentences and classify them as 'entailment' or 'not_entailment'. Language: arabic, acc: 74.00%, prompt: Can this sentence be inferred from the previous sentence? Classify it as 'entailment' or 'not_entailment'. Language: arabic, acc: 75.00%, prompt: Using a tool to analyze a conclusion, analyze the relationship between the two sentences and classify them as 'entailment' or 'not_entailment'. Language: arabic, acc: 81.00%, prompt: Is this a conclusion from the next sentence? Classify it as 'entailment' or 'not_entailment'. Language: arabic, acc: 63.00%, prompt: As part of your task in analyzing a conclusion, evaluate the relationship between the two sentences and classify them as 'entailment' or 'not_entailment' based on their relationship. Language: arabic, acc: 82.00%, prompt: Are you following this sentence directly from the previous one? Classify it as 'entailment' or 'not_entailment'. Language: spanish, acc: 78.00%, prompt: In your role as an implication analysis tool, evaluate the relationship between the given phrases and classify them as 'entailment' or 'not_entailment'. Language: spanish, acc: 73.00%, prompt: Determine whether the second sentence necessarily implies the first and label the relation as 'entailment', or as 'not_entailment' if not. Language: spanish, acc: 74.00%, prompt: Classifies the relationship between these two sentences as 'entailment' if one necessarily implies the other, or as 'not_entailment' if not. Language: spanish, acc: 80.50%, prompt: Evaluates whether the information in the second sentence is implied in the first and labels the relationship as 'entailment', or as 'not_entailment' if there is no such implication. Language: spanish, acc: 74.50%, prompt: Given a couple of phrases, label their relationship as 'entailment' if one necessarily implies the other, or as 'not_entailment' if there is no such implication. Language: spanish, acc: 77.50%, prompt: Analyzes the relationship between the phrases and classifies them as 'entailment' if one necessarily implies the other, or as 'not_entailment' if there is no such implication. Language: spanish, acc: 74.00%, prompt: Given two sentences, determine whether the second sentence is a necessary consequence of the first and label the relation as 'entailment', or as 'not_entailment' if not. Language: spanish, acc: 79.50%, prompt: Evaluates whether the information presented in the second sentence is implicit in the first and labels the relationship as 'entailment', or as 'not_entailment' if there is no such implication. Language: spanish, acc: 79.00%, prompt: Classifies the relationship between the given phrases as 'entailment' if one necessarily implies the other, or as 'not_entailment' if there is no such implication. Language: spanish, acc: 77.50%, prompt: Determines whether the information provided in the second sentence is necessarily inferable from the first and labels the relationship as 'entailment', or as 'not_entailment' if not. Language: japanese, acc: 77.50%, prompt: Analyze the relationship of a given sentence and classify it as 'entailment' or 'not_entailment'. Language: japanese, acc: 70.50%, prompt: Evaluate the semantic relationship of the sentence and classify it as 'entailment' or 'not_entailment'. Language: japanese, acc: 77.00%, prompt: Please judge the relationship between the given sentences and classify them as 'entailment' or 'not_entailment'. Language: japanese, acc: 77.50%, prompt: Examine the paraphrases of a given sentence and classify them 'entailment' or 'not_entailment'. Language: japanese, acc: 53.50%, prompt: Rate the similarity of a given sentence and categorize it as 'entailment' or 'not_entailment'. Language: japanese, acc: 54.00%, prompt: Determinate the semantic connections of a given sentence and classify it as 'entailment' or 'not_entailment'. Language: japanese, acc: 81.00%, prompt: Examine the semantic match of a given sentence and categorize it as 'entailment' or 'not_entailment'. Language: japanese, acc: 79.00%, prompt: Classify it as 'entailment' or 'not_entailment' based on the content of the sentence. Language: japanese, acc: 78.50%, prompt: Analyze the relationship of a given sentence and classify it as 'entailment' or 'not_entailment'. Language: japanese, acc: 79.50%, prompt: Judge the semantic connections of a given sentence and categorize it as 'entailment' or 'not_entailment'. Language: korean, acc: 76.50%, prompt: Evaluate the relationship between any two sentences given to you and classify you as 'entailment' or 'not_entailment'. Language: korean, acc: 78.00%, prompt: Analyze the semantic deductive relations between sentences and classify them as 'entailment' or 'not_entailment'. Language: korean, acc: 78.00%, prompt: Evaluate the logical relevance between sentences and classify them as 'entailment' or 'not_entailment'. Language: korean, acc: 80.50%, prompt: Evaluate the interaction of two given sentences and classify them as 'entailment' or 'not_entailment'. Language: korean, acc: 82.00%, prompt: Please check whether there is a semantic match between those two sentences and classify them as 'entailment' or 'not_entailment'. Language: korean, acc: 78.50%, prompt: Compare information between given sentences and classify them as 'entailment' or 'not_entailment'. Language: korean, acc: 57.00%, prompt: Please analyse the correlation between those two sentences and classify them as 'entailment' or 'not_entailment'. Language: korean, acc: 77.00%, prompt: Evaluate the different meanings between given sentences and classify them as 'entailment' or 'not_entailment'. Language: korean, acc: 54.00%, prompt: Compare the semantic structure of the two sentences and classify them as 'entailment' or 'not_entailment'. Language: korean, acc: 76.50%, prompt: Evaluate the interactions between sentences and classify them as 'entailment' or 'not_entailment'. # sst2 ## 10 prompts Acc: 96.00%, prompt: Read the provided excerpt and choose between 'positive' and 'negative' to describe its sentiment: Acc: 93.00%, prompt: Determine the overall sentiment of this sentence, categorizing it as 'positive' or 'negative': Acc: 90.50%, prompt: Review this statement and decide whether it has a 'positive' or 'negative' sentiment: Acc: 90.00%, prompt: Analyze the tone of this statement and respond with either 'positive' or 'negative': Acc: 90.00%, prompt: Please identify the emotional tone of this passage: 'positive' or 'negative'? Acc: 88.50%, prompt: Evaluate the sentiment of the given text and classify it as 'positive' or 'negative': Acc: 88.50%, prompt: Assess the mood of the following quote and determine if it's 'positive' or 'negative': Acc: 88.00%, prompt: After examining the following expression, label its emotion as either 'positive' or 'negative': Acc: 87.50%, prompt: Given the context of this text, indicate if the emotion conveyed is 'positive' or 'negative': Acc: 86.00%, prompt: Considering the given phrase, would you say it carries a 'positive' or 'negative' connotation? Acc: 94.00%, prompt: Emulating a sentiment classification system, indicate whether the provided text is 'positive' or 'negative'. Classify: Acc: 93.50%, prompt: Functioning as a sentiment identification tool, assess if the following expression is 'positive' or 'negative'. Classify: Acc: 93.00%, prompt: Acting as a sentiment evaluator, identify if the given sentence is 'positive' or 'negative'. Classify: Acc: 92.50%, prompt: As a sentiment classifier, determine whether the following text is 'positive' or 'negative'. Please classify: Acc: 92.50%, prompt: Working as a sentiment analyzer, please indicate if the following text is 'positive' or 'negative'. Classify: Acc: 92.00%, prompt: In the role of a sentiment analysis tool, respond with 'positive' or 'negative' to classify this statement: Acc: 92.00%, prompt: Serving as a sentiment evaluation model, determine if the given statement is 'positive' or 'negative'. Classify: Acc: 90.50%, prompt: In the capacity of a sentiment classifier, decide whether the given quote is 'positive' or 'negative'. Classify: Acc: 90.00%, prompt: Taking on the role of an emotion classifier, specify if the provided phrase is 'positive' or 'negative'. Classify: Acc: 88.50%, prompt: As an emotion detector, determine if the provided passage conveys a 'positive' or 'negative' sentiment. Classify: ## bertattack Original prompt: Read the provided excerpt and choose between 'positive' and 'negative' to describe its sentiment: Attacked prompt: b"read the standard excerpt and pause between 'positive' and 'negative' to describe its thoughtful: " Original acc: 96.00%, attacked acc: 2.00%, dropped acc: 94.00% Original prompt: Determine the overall sentiment of this sentence, categorizing it as 'positive' or 'negative': Attacked prompt: b"ve the overall sentiment of this situation, categorizing it as 'positive' or 'negative': " Original acc: 93.00%, attacked acc: 86.00%, dropped acc: 7.00% Original prompt: Review this statement and decide whether it has a 'positive' or 'negative' sentiment: Attacked prompt: b"what this assertion and are whether it has a 'positive' or 'negative' sentiment: " Original acc: 90.00%, attacked acc: 0.00%, dropped acc: 90.00% Original prompt: Acting as a sentiment evaluator, identify if the given sentence is 'positive' or 'negative'. Classify: Attacked prompt: b"on as a sentiment evaluator, identify if the given event is 'positive' or 'negative'. Classify: " Original acc: 93.00%, attacked acc: 85.00%, dropped acc: 8.00% Original prompt: Emulating a sentiment classification system, indicate whether the provided text is 'positive' or 'negative'. Classify: Attacked prompt: b".if a sentiment index system, indicate perhaps the created one is 'positive' or 'negative'. Classify: " Original acc: 94.00%, attacked acc: 80.50%, dropped acc: 13.50% Original prompt: Functioning as a sentiment identification tool, assess if the following expression is 'positive' or 'negative'. Classify: Attacked prompt: b"from as a sentiments identification tool, assess if the same word is 'positive' or 'negative'. function: " Original acc: 93.00%, attacked acc: 76.00%, dropped acc: 17.00% ## checklist Original prompt: Read the provided excerpt and choose between 'positive' and 'negative' to describe its sentiment: Attacked prompt: b"Read the provided excerpt and choose between 'positive' and 'negative' to describe its sentiment Q47Jmd4lMV: " Original acc: 96.00%, attacked acc: 94.50%, dropped acc: 1.50% Original prompt: Determine the overall sentiment of this sentence, categorizing it as 'positive' or 'negative': Attacked prompt: b"Determine the overall sentiment of this sentence, categorizing it as 'positive' or 'negative' AMsRIKZniY: " Original acc: 93.00%, attacked acc: 90.00%, dropped acc: 3.00% Original prompt: Review this statement and decide whether it has a 'positive' or 'negative' sentiment: Attacked prompt: b"Review this statement and decide whether it has a 'positive' or 'negative' sentiment ofw9fEkN5R: " Original acc: 91.00%, attacked acc: 90.00%, dropped acc: 1.00% Original prompt: Emulating a sentiment classification system, indicate whether the provided text is 'positive' or 'negative'. Classify: Attacked prompt: b"Emulating a sentiment classification system, indicate whether the provided text is 'positive' or 'negative'. Classify PDCGfz6DL9: " Original acc: 94.00%, attacked acc: 93.00%, dropped acc: 1.00% Original prompt: Acting as a sentiment evaluator, identify if the given sentence is 'positive' or 'negative'. Classify: Attacked prompt: b"Acting as a sentiment evaluator, identify if the given sentence is 'positive' or 'negative'. Classify Q47Jmd4lMV: " Original acc: 93.50%, attacked acc: 93.00%, dropped acc: 0.50% Original prompt: Functioning as a sentiment identification tool, assess if the following expression is 'positive' or 'negative'. Classify: Attacked prompt: b"Functioning as a sentiment identification tool, assess if the following expression is 'positive' or 'negative'. Classify Cev0s49fbe: " Original acc: 92.50%, attacked acc: 92.00%, dropped acc: 0.50% ## deepwordbug Original prompt: Read the provided excerpt and choose between 'positive' and 'negative' to describe its sentiment: Attacked prompt: b"Read the provded excertp and choost between 'positive' and 'negative' to descrhibe its sentigment: " Original acc: 95.50%, attacked acc: 91.00%, dropped acc: 4.50% Original prompt: Determine the overall sentiment of this sentence, categorizing it as 'positive' or 'negative': Attacked prompt: b"Deetrmine the overall sentLment of this sentence, categorizing it as 'positive' or 'negative': " Original acc: 93.00%, attacked acc: 88.50%, dropped acc: 4.50% Original prompt: Review this statement and decide whether it has a 'positive' or 'negative' sentiment: Attacked prompt: b"Review this statement and decide whether it has a 'positive' or 'negative' sentiment: " Original acc: 90.50%, attacked acc: 90.50%, dropped acc: 2.50% Original prompt: Emulating a sentiment classification system, indicate whether the provided text is 'positive' or 'negative'. Classify: Attacked prompt: b"Emulating a seyntiment classifciation ssytem, indicate wJhether the provided tLext is 'positive' or 'negative'. Classify: " Original acc: 94.00%, attacked acc: 88.50%, dropped acc: 5.50% Original prompt: Functioning as a sentiment identification tool, assess if the following expression is 'positive' or 'negative'. Classify: Attacked prompt: b"uFnctioning as a sentmient identification tool, assess if the folGowing expression is 'positive' or 'negative'. Classify: " Original acc: 93.00%, attacked acc: 85.00%, dropped acc: 8.00% Original prompt: Acting as a sentiment evaluator, identify if the given sentence is 'positive' or 'negative'. Classify: Attacked prompt: b"Actign as a sentiemnt evaluator, Qdentify if the liven seXtence is 'positive' or 'negative'. ClasOify: " Original acc: 93.00%, attacked acc: 87.00%, dropped acc: 6.00% ## stresstest Original prompt: Read the provided excerpt and choose between 'positive' and 'negative' to describe its sentiment: Attacked prompt: b"Read the provided excerpt and choose between 'positive' and 'negative' to describe its sentiment and false is not true : " Original acc: 95.50%, attacked acc: 96.50%, dropped acc: -1.00% Original prompt: Determine the overall sentiment of this sentence, categorizing it as 'positive' or 'negative': Attacked prompt: b"Determine the overall sentiment of this sentence, categorizing it as 'positive' or 'negative' and false is not true : " Original acc: 93.00%, attacked acc: 93.50%, dropped acc: -0.50% Original prompt: Review this statement and decide whether it has a 'positive' or 'negative' sentiment: Attacked prompt: b"Review this statement and decide whether it has a 'positive' or 'negative' sentiment and true is true : " Original acc: 91.00%, attacked acc: 0.00%, dropped acc: 91.00% Original prompt: Functioning as a sentiment identification tool, assess if the following expression is 'positive' or 'negative'. Classify: Attacked prompt: b"Functioning as a sentiment identification tool, assess if the following expression is 'positive' or 'negative'. Classify and false is not true : " Original acc: 93.50%, attacked acc: 93.00%, dropped acc: 0.50% Original prompt: Emulating a sentiment classification system, indicate whether the provided text is 'positive' or 'negative'. Classify: Attacked prompt: b"Emulating a sentiment classification system, indicate whether the provided text is 'positive' or 'negative'. Classify and false is not true : " Original acc: 94.00%, attacked acc: 94.50%, dropped acc: -0.50% Original prompt: Acting as a sentiment evaluator, identify if the given sentence is 'positive' or 'negative'. Classify: Attacked prompt: b"Acting as a sentiment evaluator, identify if the given sentence is 'positive' or 'negative'. Classify and false is not true : " Original acc: 93.50%, attacked acc: 94.00%, dropped acc: -0.50% ## textbugger Original prompt: Read the provided excerpt and choose between 'positive' and 'negative' to describe its sentiment: Attacked prompt: b"Read the yielded excerpt and chosen between 'positive' and 'negative' to describe its mood: " Original acc: 95.50%, attacked acc: 89.50%, dropped acc: 6.00% Original prompt: Determine the overall sentiment of this sentence, categorizing it as 'positive' or 'negative': Attacked prompt: b"Determining the ov erall vibe of this sentence, categorizing it as 'positive' or 'negative': " Original acc: 93.00%, attacked acc: 85.50%, dropped acc: 7.50% Original prompt: Review this statement and decide whether it has a 'positive' or 'negative' sentiment: Attacked prompt: b"Review this speeches and decide if it has a 'positive' or 'negative' sentiment: " Original acc: 90.00%, attacked acc: 82.50%, dropped acc: 7.50% Original prompt: Emulating a sentiment classification system, indicate whether the provided text is 'positive' or 'negative'. Classify: Attacked prompt: b"Emulating a sentiment classifciation system, indicating \xd4\x9dhether the provided text is 'positive' or 'negative'. Categorize: " Original acc: 93.50%, attacked acc: 89.00%, dropped acc: 4.50% Original prompt: Acting as a sentiment evaluator, identify if the given sentence is 'positive' or 'negative'. Classify: Attacked prompt: b"Atcing as a sentiemnt evalua\xf0\x9d\x9a\x9dor, detecting if the given sentence is 'positive' or 'negative'. Classify: " Original acc: 93.50%, attacked acc: 86.00%, dropped acc: 7.50% Original prompt: Functioning as a sentiment identification tool, assess if the following expression is 'positive' or 'negative'. Classify: Attacked prompt: b"Functio\xd5\xb8ing as a sentiment identification tools, assess if the later ex\xd1\x80ression is 'positive' or 'negative'. Classify: " Original acc: 93.50%, attacked acc: 86.50%, dropped acc: 7.00% ## textfooler Original prompt: Read the provided excerpt and choose between 'positive' and 'negative' to describe its sentiment: Attacked prompt: b"Read the lent excerpt and akhtar between 'positive' and 'negative' to schematic its senses: " Original acc: 96.00%, attacked acc: 47.50%, dropped acc: 48.50% Original prompt: Determine the overall sentiment of this sentence, categorizing it as 'positive' or 'negative': Attacked prompt: b"Define the unmitigated foreboding of this sanctions, categorizing it as 'positive' or 'negative': " Original acc: 92.50%, attacked acc: 51.00%, dropped acc: 41.50% Original prompt: Review this statement and decide whether it has a 'positive' or 'negative' sentiment: Attacked prompt: b"Analysis this assertion and defining whether it ai a 'positive' or 'negative' sentiment: " Original acc: 90.50%, attacked acc: 0.00%, dropped acc: 90.50% Original prompt: Emulating a sentiment classification system, indicate whether the provided text is 'positive' or 'negative'. Classify: Attacked prompt: b"Emulating a foreboding classifies regime, explain whether the stipulated laws is 'positive' or 'negative'. Grading: " Original acc: 94.00%, attacked acc: 4.50%, dropped acc: 89.50% Original prompt: Acting as a sentiment evaluator, identify if the given sentence is 'positive' or 'negative'. Classify: Attacked prompt: b"Acting as a feeling evaluator, defining if the given sorrows is 'positive' or 'negative'. Categorize: " Original acc: 93.00%, attacked acc: 78.50%, dropped acc: 14.50% Original prompt: Functioning as a sentiment identification tool, assess if the following expression is 'positive' or 'negative'. Classify: Attacked prompt: b"Exploitation as a sentiment definitions device, valuation if the farther demonstration is 'positive' or 'negative'. Categories: " Original acc: 92.00%, attacked acc: 7.50%, dropped acc: 84.50% ## translation Language: chinese, acc: 95.50%, prompt: Answer whether the statement is 'positive' or 'negative' based on sentiment analysis. Language: chinese, acc: 90.50%, prompt: As an emotion analysis tool, determine whether the emotion in the text is 'positive' or 'negative'. Language: chinese, acc: 92.00%, prompt: Categorize the statement as 'positive' or 'negative', based on its emotional bent. Language: chinese, acc: 91.50%, prompt: Please use sentiment analysis to classify the text as 'positive' or 'negative'. Language: chinese, acc: 91.00%, prompt: Please determine whether the emotion of the sentence is 'positive' or 'negative' and categorize it. Language: chinese, acc: 92.00%, prompt: Using sentiment analysis, classify the text as 'positive' or 'negative'. Language: chinese, acc: 91.00%, prompt: Please answer whether the emotion of the sentence is 'positive' or 'negative' and categorize it. Language: chinese, acc: 93.50%, prompt: Categorize the statement as 'positive' or 'negative' based on sentiment analysis. Language: chinese, acc: 89.50%, prompt: Please judge this text as 'positive' or 'negative' according to the criteria of sentiment analysis. Language: chinese, acc: 89.00%, prompt: Please classify this sentence as 'positive' or 'negative' according to the principles of emotion analysis. Language: french, acc: 94.50%, prompt: As a sentiment analysis tool, please answer with 'positive' or 'negative' to classify this statement. Language: french, acc: 91.50%, prompt: Determine whether this phrase is 'positive' or 'negative' as a sentiment classification tool. Language: french, acc: 94.00%, prompt: Identify the tone of this statement by choosing between 'positive' and 'negative' as a sentiment analysis tool. Language: french, acc: 90.50%, prompt: Use sentiment analysis to classify this statement as 'positive' or 'negative'. Language: french, acc: 92.50%, prompt: As a sentiment classification tool, please determine whether this statement is 'positive' or 'negative'. Language: french, acc: 93.00%, prompt: Classify this sentence as 'positive' or 'negative' using sentiment analysis. Language: french, acc: 93.50%, prompt: Choose between 'positive' or 'negative' to classify this statement as a sentiment analysis tool. Language: french, acc: 93.50%, prompt: Identify the sentiment expressed in this statement by selecting 'positive' or 'negative' as a sentiment classification tool. Language: french, acc: 92.00%, prompt: Determine whether this phrase is 'positive' or 'negative' using sentiment analysis as a classification tool. Language: french, acc: 90.50%, prompt: Use sentiment analysis to classify this statement as 'positive' or 'negative'. Language: arabic, acc: 90.00%, prompt: Under emotional analysis, answer 'positive' or 'negative' to classify this statement. Language: arabic, acc: 94.00%, prompt: Does this statement express a 'positive' or 'negative' reaction? Language: arabic, acc: 90.50%, prompt: Is that a 'positive' or a 'negative' phrase? Language: arabic, acc: 52.50%, prompt: What is the classification between 'positive' and 'negative'? Language: arabic, acc: 89.00%, prompt: Does this sentence express 'positive' or 'negative' feelings? Language: arabic, acc: 29.50%, prompt: In the context of textual analysis, what classification is this phrase between 'positive' and 'negative'? Language: arabic, acc: 89.00%, prompt: Could this be classified as 'positive' or 'negative'? Language: arabic, acc: 80.50%, prompt: In the context of emotional analysis, what classification is this statement between 'positive' and 'negative'? Language: arabic, acc: 86.50%, prompt: Can this be classified as 'positive' or 'negative'? Language: arabic, acc: 88.00%, prompt: Under the classification of emotions, is this sentence 'positive' or 'negative'? Language: spanish, acc: 90.00%, prompt: As a feeling analysis tool, classify this statement as 'positive' or 'negative'. Language: spanish, acc: 90.50%, prompt: Determine whether this statement has a 'positive' or 'negative' connotation. Language: spanish, acc: 96.50%, prompt: Indicate whether the following statement is 'positive' or 'negative'. Language: spanish, acc: 88.00%, prompt: Evaluate whether this text has a 'positive' or 'negative' emotional charge. Language: spanish, acc: 93.50%, prompt: According to your sentiment analysis, would you say this comment is 'positive' or 'negative'? Language: spanish, acc: 94.00%, prompt: In the context of sentiment analysis, label this sentence as 'positive' or 'negative'. Language: spanish, acc: 95.00%, prompt: Rate the following statement as 'positive' or 'negative', according to your sentiment analysis. Language: spanish, acc: 88.50%, prompt: How would you classify this text in terms of its emotional tone? 'positive' or 'negative'? Language: spanish, acc: 89.00%, prompt: As a tool for sentiment analysis, would you say this statement is 'positive' or 'negative'? Language: spanish, acc: 92.50%, prompt: Classify this statement as 'positive' or 'negative', please. Language: japanese, acc: 92.00%, prompt: Treat this sentence as an emotion analysis tool and categorize it as 'positive' and 'negative'. Language: japanese, acc: 88.50%, prompt: Use this article as a sentiment analysis tool to classify 'positive' and 'negative'. Language: japanese, acc: 88.00%, prompt: Use this sentence as an emotion analysis tool to determine whether it is 'positive' or 'negative'. Language: japanese, acc: 92.00%, prompt: Use this sentence as an emotion analysis tool to classify 'positive' and 'negative'. Language: japanese, acc: 89.00%, prompt: Use this sentence as a sentiment analysis tool and classify it as 'positive' or 'negative'. Language: japanese, acc: 87.50%, prompt: To classify this sentence as 'positive' or 'negative', evaluate it as a sentiment analysis tool. Language: japanese, acc: 89.50%, prompt: Treat this sentence as an emotion analysis tool to determine whether it is 'positive' or 'negative'. Language: japanese, acc: 89.00%, prompt: Use this sentence as a sentiment analysis tool to classify 'positive' and 'negative'. Language: japanese, acc: 87.00%, prompt: Analyze this sentence as an emotion analysis tool to classify whether it is 'positive' or 'negative'. Language: japanese, acc: 87.50%, prompt: Use this sentence as an emotional analysis tool to determine whether it is 'positive' or 'negative'. Language: korean, acc: 93.50%, prompt: As an emotional analysis tool, respond with 'positive' or 'negative' to classify these sentences. Language: korean, acc: 93.00%, prompt: Classify this sentence as 'positive' if you regard it as positive, 'negative' if you regard it as negative. Language: korean, acc: 85.00%, prompt: Please rate the emotion of this sentence and classify it as 'positive' or 'negative'. Language: korean, acc: 93.50%, prompt: Classify this sentence as 'positive' if you perceive it positively and 'negative' if you perceive it negatively. Language: korean, acc: 94.50%, prompt: If this is a sentence delivered using a positive expression, classify it as 'positive' and if this is a sentence delivered using a negative expression, classify it as 'negative'. Language: korean, acc: 95.00%, prompt: Respond with 'positive' or 'negative' by categorizing whether the sentence is positive or negative. Language: korean, acc: 87.50%, prompt: Please analyze the emotion in this sentence and classify it as 'positive' or 'negative'. Language: korean, acc: 95.50%, prompt: Classify this sentence as 'positive' if it contains a positive meaning, 'negative' if it contains a negative meaning. Language: korean, acc: 94.50%, prompt: Classify this sentence as 'positive' if it contains positive content, 'negative' if it contains negative content. Language: korean, acc: 89.50%, prompt: Classify what you're trying to convey in this sentence as 'positive' if it's positive, and 'negative' if it's negative. # wnli ## 10 prompts Acc: 67.61%, prompt: Examine the pair of sentences and determine if they exhibit entailment or not_entailment. Answer with either 'entailment' or 'not_entailment'. Acc: 67.61%, prompt: Identify whether the given pair of sentences demonstrates entailment or not_entailment. Answer with 'entailment' or 'not_entailment'. Acc: 64.79%, prompt: Are the following two sentences entailment or not_entailment? Answer me with "entailment" or "not_entailment", just one word. Acc: 63.38%, prompt: Determine if the given pair of sentences displays entailment or not_entailment. Respond with 'entailment' or 'not_entailment'. Acc: 59.15%, prompt: Does the relationship between the given sentences represent entailment or not_entailment? Respond with 'entailment' or 'not_entailment'. Acc: 56.34%, prompt: Analyze the two provided sentences and decide if their relationship is 'entailment' or 'not_entailment'. Acc: 54.93%, prompt: Review the two statements and categorize their relationship as either 'entailment' or 'not_entailment'. Acc: 52.11%, prompt: Assess the connection between the following sentences and classify it as 'entailment' or 'not_entailment'. Acc: 52.11%, prompt: Indicate if the connection between the following sentences is 'entailment' or 'not_entailment'. Acc: 50.70%, prompt: Please classify the relationship between the provided sentences as 'entailment' or 'not_entailment'. Acc: 50.70%, prompt: Considering the two sentences, identify if their relationship is 'entailment' or 'not_entailment'. Acc: 66.20%, prompt: Acting as an entailment detection instrument, determine if the given pair of sentences demonstrates entailment or not_entailment. Answer with 'entailment' or 'not_entailment': Acc: 63.38%, prompt: In the role of a semantic relationship analyzer, examine the connection between the given sentences and decide if they exhibit entailment or not_entailment. Answer with 'entailment' or 'not_entailment': Acc: 59.15%, prompt: Working as an entailment classifier, identify whether the given pair of sentences displays entailment or not_entailment. Respond with 'entailment' or 'not_entailment': Acc: 57.75%, prompt: As a tool for determining entailment relationships, review the two statements and categorize their connection as either 'entailment' or 'not_entailment': Acc: 57.75%, prompt: In the capacity of an entailment assessment system, indicate if the link between the following sentences is 'entailment' or 'not_entailment': Acc: 57.75%, prompt: As an instrument for entailment evaluation, consider the two sentences and determine if their relationship is 'entailment' or 'not_entailment'. Respond with 'entailment' or 'not_entailment': Acc: 52.11%, prompt: In your role as an entailment analysis tool, assess the relationship between the given sentences and classify it as 'entailment' or 'not_entailment': Acc: 50.70%, prompt: As an entailment identification system, examine the connection between the following sentences and respond with 'entailment' or 'not_entailment': Acc: 50.70%, prompt: While performing entailment analysis, classify the relationship between the provided sentences as 'entailment' or 'not_entailment': Acc: 47.89%, prompt: Functioning as an entailment evaluation tool, analyze the provided sentences and decide if their relationship is 'entailment' or 'not_entailment': ## bertattack Original prompt: Identify whether the given pair of sentences demonstrates entailment or not_entailment. Answer with 'entailment' or 'not_entailment'. Attacked prompt: b"Identify whether the given individual of verbs demonstrates entailment or not_entailment. have with 'entailment' or 'not_entailment'." Original acc: 69.01%, attacked acc: 47.89%, dropped acc: 21.13% Original prompt: Examine the pair of sentences and determine if they exhibit entailment or not_entailment. Answer with either 'entailment' or 'not_entailment'. Attacked prompt: b"Examine the 2 of pronouns and determine if they exhibit entailment or not_entailment. Answer with a 'entailment' or 'not_entailment'." Original acc: 66.20%, attacked acc: 49.30%, dropped acc: 16.90% Original prompt: Are the following two sentences entailment or not_entailment? Answer me with "entailment" or "not_entailment", just one word. Attacked prompt: b'Are the after between sentences entailment or not_entailment? Answer me with "entailment" or "not_entailment", just first two. ' Original acc: 63.38%, attacked acc: 50.70%, dropped acc: 12.68% Original prompt: Acting as an entailment detection instrument, determine if the given pair of sentences demonstrates entailment or not_entailment. Answer with 'entailment' or 'not_entailment': Attacked prompt: b"her as an entailment detection instrument, determine if the given case of sentences provides entailment or not_entailment. example with 'entailment' or 'not_entailment':" Original acc: 67.61%, attacked acc: 49.30%, dropped acc: 18.31% Original prompt: In the role of a semantic relationship analyzer, examine the connection between the given sentences and decide if they exhibit entailment or not_entailment. Answer with 'entailment' or 'not_entailment': Attacked prompt: b"her the role of a semantic relationship analyzer, examine the connection between the input sentences and decide if they appear entailment or not_entailment. get with 'entailment' or 'not_entailment':" Original acc: 63.38%, attacked acc: 50.70%, dropped acc: 12.68% Original prompt: Working as an entailment classifier, identify whether the given pair of sentences displays entailment or not_entailment. Respond with 'entailment' or 'not_entailment': Attacked prompt: b"Working as an entailment classifier, identify whether the given corpus of verbs displays entailment or not_entailment. act with 'entailment' or 'not_entailment':" Original acc: 60.56%, attacked acc: 47.89%, dropped acc: 12.68% ## checklist Original prompt: Identify whether the given pair of sentences demonstrates entailment or not_entailment. Answer with 'entailment' or 'not_entailment'. Attacked prompt: b"Identify whether the given pair of sentences demonstrates entailment or not_entailment. Answer with 'entailment' or 'not_entailment' 6a4Yn3RGVc." Original acc: 69.01%, attacked acc: 67.61%, dropped acc: 1.41% Original prompt: Are the following two sentences entailment or not_entailment? Answer me with "entailment" or "not_entailment", just one word. Attacked prompt: b'Are the following two sentences entailment or not_entailment? Answer me with "entailment" or "not_entailment", just one word zexKUTWJYG. ' Original acc: 64.79%, attacked acc: 59.15%, dropped acc: 5.63% Original prompt: Examine the pair of sentences and determine if they exhibit entailment or not_entailment. Answer with either 'entailment' or 'not_entailment'. Attacked prompt: b"Examine the pair of sentences and determine if they exhibit entailment or not_entailment. Answer with either 'entailment' or 'not_entailment' 5XeflW1ZJc." Original acc: 67.61%, attacked acc: 61.97%, dropped acc: 5.63% Original prompt: Acting as an entailment detection instrument, determine if the given pair of sentences demonstrates entailment or not_entailment. Answer with 'entailment' or 'not_entailment': Attacked prompt: b"Acting as an entailment detection instrument, determine if the given pair of sentences demonstrates entailment or not_entailment. Answer with 'entailment' or 'not_entailment' 5XeflW1ZJc:" Original acc: 67.61%, attacked acc: 64.79%, dropped acc: 2.82% Original prompt: In the role of a semantic relationship analyzer, examine the connection between the given sentences and decide if they exhibit entailment or not_entailment. Answer with 'entailment' or 'not_entailment': Attacked prompt: b"In the role of a semantic relationship analyzer, examine the connection between the given sentences and decide if they exhibit entailment or not_entailment. Answer with 'entailment' or 'not_entailment' 5XeflW1ZJc:" Original acc: 63.38%, attacked acc: 57.75%, dropped acc: 5.63% Original prompt: Working as an entailment classifier, identify whether the given pair of sentences displays entailment or not_entailment. Respond with 'entailment' or 'not_entailment': Attacked prompt: b"Working as an entailment classifier, identify whether the given pair of sentences displays entailment or not_entailment. Respond with 'entailment' or 'not_entailment' is77sOXAu8:" Original acc: 60.56%, attacked acc: 59.15%, dropped acc: 1.41% ## deepwordbug Original prompt: Identify whether the given pair of sentences demonstrates entailment or not_entailment. Answer with 'entailment' or 'not_entailment'. Attacked prompt: b"Identify whether the given paiM of senJences demonstartes entailment or not_entailment. Answer with 'entailment' or 'not_entailment'." Original acc: 69.01%, attacked acc: 57.75%, dropped acc: 11.27% Original prompt: Examine the pair of sentences and determine if they exhibit entailment or not_entailment. Answer with either 'entailment' or 'not_entailment'. Attacked prompt: b"Examine the Bair of snetences and determine if they exhibit entailment or not_entailment. Ancwer with eMther 'entailment' or 'not_entailment'." Original acc: 66.20%, attacked acc: 53.52%, dropped acc: 12.68% Original prompt: Are the following two sentences entailment or not_entailment? Answer me with "entailment" or "not_entailment", just one word. Attacked prompt: b'Are the following wo sentences entailment or not_entailment? Answer me with "entailment" or "not_entailment", just one word. ' Original acc: 66.20%, attacked acc: 57.75%, dropped acc: 8.45% Original prompt: Acting as an entailment detection instrument, determine if the given pair of sentences demonstrates entailment or not_entailment. Answer with 'entailment' or 'not_entailment': Attacked prompt: b"Acting as an entailment detection instrument, deterimne if the given pari of sentences demonstrates entailment or not_entailment. AnsNer with 'entailment' or 'not_entailment':" Original acc: 67.61%, attacked acc: 54.93%, dropped acc: 12.68% Original prompt: In the role of a semantic relationship analyzer, examine the connection between the given sentences and decide if they exhibit entailment or not_entailment. Answer with 'entailment' or 'not_entailment': Attacked prompt: b"IQn the role of a semantic relationship analyzer, eamine the connection between the given sentences and decide if they exhbit entailment or not_entailment. Answer with 'entailment' or 'not_entailment':" Original acc: 63.38%, attacked acc: 53.52%, dropped acc: 9.86% Original prompt: Working as an entailment classifier, identify whether the given pair of sentences displays entailment or not_entailment. Respond with 'entailment' or 'not_entailment': Attacked prompt: b"Working as an entailment classifier, identify whether the given pair of sentences displays entailment or not_entailment. Respond with 'entailment' or 'not_entailment':" Original acc: 59.15%, attacked acc: 59.15%, dropped acc: 4.23% ## stresstest Original prompt: Identify whether the given pair of sentences demonstrates entailment or not_entailment. Answer with 'entailment' or 'not_entailment'. Attacked prompt: b"Identify whether the given pair of sentences demonstrates entailment or not_entailment. Answer with 'entailment' or 'not_entailment' and false is not true ." Original acc: 70.42%, attacked acc: 66.20%, dropped acc: 4.23% Original prompt: Examine the pair of sentences and determine if they exhibit entailment or not_entailment. Answer with either 'entailment' or 'not_entailment'. Attacked prompt: b"Examine the pair of sentences and determine if they exhibit entailment or not_entailment. Answer with either 'entailment' or 'not_entailment' and false is not true ." Original acc: 67.61%, attacked acc: 64.79%, dropped acc: 2.82% Original prompt: Are the following two sentences entailment or not_entailment? Answer me with "entailment" or "not_entailment", just one word. Attacked prompt: b'Are the following two sentences entailment or not_entailment? Answer me with "entailment" or "not_entailment", just one word and true is true . ' Original acc: 61.97%, attacked acc: 63.38%, dropped acc: -1.41% Original prompt: Acting as an entailment detection instrument, determine if the given pair of sentences demonstrates entailment or not_entailment. Answer with 'entailment' or 'not_entailment': Attacked prompt: b"Acting as an entailment detection instrument, determine if the given pair of sentences demonstrates entailment or not_entailment. Answer with 'entailment' or 'not_entailment' and true is true :" Original acc: 66.20%, attacked acc: 64.79%, dropped acc: 1.41% Original prompt: In the role of a semantic relationship analyzer, examine the connection between the given sentences and decide if they exhibit entailment or not_entailment. Answer with 'entailment' or 'not_entailment': Attacked prompt: b"In the role of a semantic relationship analyzer, examine the connection between the given sentences and decide if they exhibit entailment or not_entailment. Answer with 'entailment' or 'not_entailment' and true is true and true is true and true is true and true is true and true is true :" Original acc: 63.38%, attacked acc: 59.15%, dropped acc: 4.23% Original prompt: Working as an entailment classifier, identify whether the given pair of sentences displays entailment or not_entailment. Respond with 'entailment' or 'not_entailment': Attacked prompt: b"Working as an entailment classifier, identify whether the given pair of sentences displays entailment or not_entailment. Respond with 'entailment' or 'not_entailment' and false is not true :" Original acc: 60.56%, attacked acc: 59.15%, dropped acc: 1.41% ## textbugger Original prompt: Examine the pair of sentences and determine if they exhibit entailment or not_entailment. Answer with either 'entailment' or 'not_entailment'. Attacked prompt: b"Examine the pai\xe2\xb2\x85 of sentences and determines if they exhibit entailment or not_entailment. Answer with neither 'entailment' or 'not_entailment'." Original acc: 67.61%, attacked acc: 52.11%, dropped acc: 15.49% Original prompt: Identify whether the given pair of sentences demonstrates entailment or not_entailment. Answer with 'entailment' or 'not_entailment'. Attacked prompt: b"Identify whether the given pairs of sentence demonst rates entailment or not_entailment. Answering with 'entailment' or 'not_entailment'." Original acc: 69.01%, attacked acc: 56.34%, dropped acc: 12.68% Original prompt: Are the following two sentences entailment or not_entailment? Answer me with "entailment" or "not_entailment", just one word. Attacked prompt: b'Are the following two sentences entailment or not_entailment? Answer me with "entailment" or "not_entailment", just one word. ' Original acc: 63.38%, attacked acc: 63.38%, dropped acc: 5.63% Original prompt: Acting as an entailment detection instrument, determine if the given pair of sentences demonstrates entailment or not_entailment. Answer with 'entailment' or 'not_entailment': Attacked prompt: b"Acting as an entailment detection instrument, determine if the given couple of sentences demonstrates entailment or not_entailment. Answering with 'entailment' or 'not_entailment':" Original acc: 66.20%, attacked acc: 54.93%, dropped acc: 11.27% Original prompt: In the role of a semantic relationship analyzer, examine the connection between the given sentences and decide if they exhibit entailment or not_entailment. Answer with 'entailment' or 'not_entailment': Attacked prompt: b"Onto the role of a semantic relaionship analyzer, examine the connection between the given sentences and decide if they exposition entailment or not_entailment. Answer with 'entailment' or 'not_entailment':" Original acc: 63.38%, attacked acc: 53.52%, dropped acc: 9.86% Original prompt: Working as an entailment classifier, identify whether the given pair of sentences displays entailment or not_entailment. Respond with 'entailment' or 'not_entailment': Attacked prompt: b"Working as an entailment classifier, detects whether the given pair of sentences displays entailment or not_entailment. Answering with 'entailment' or 'not_entailment':" Original acc: 60.56%, attacked acc: 52.11%, dropped acc: 8.45% ## textfooler Original prompt: Identify whether the given pair of sentences demonstrates entailment or not_entailment. Answer with 'entailment' or 'not_entailment'. Attacked prompt: b"Identify whether the given pair of sentence exposes entailment or not_entailment. Retaliated with 'entailment' or 'not_entailment'." Original acc: 69.01%, attacked acc: 52.11%, dropped acc: 16.90% Original prompt: Are the following two sentences entailment or not_entailment? Answer me with "entailment" or "not_entailment", just one word. Attacked prompt: b'Are the following two sentences entailment or not_entailment? Answer me with "entailment" or "not_entailment", just one word. ' Original acc: 63.38%, attacked acc: 63.38%, dropped acc: 5.63% Original prompt: Examine the pair of sentences and determine if they exhibit entailment or not_entailment. Answer with either 'entailment' or 'not_entailment'. Attacked prompt: b"Examine the peers of sentences and determine if they exhibit entailment or not_entailment. Answer with either 'entailment' or 'not_entailment'." Original acc: 67.61%, attacked acc: 53.52%, dropped acc: 14.08% Original prompt: Acting as an entailment detection instrument, determine if the given pair of sentences demonstrates entailment or not_entailment. Answer with 'entailment' or 'not_entailment': Attacked prompt: b"Acting as an entailment detection instrument, determine if the given torque of sentences depicting entailment or not_entailment. Reactions with 'entailment' or 'not_entailment':" Original acc: 67.61%, attacked acc: 46.48%, dropped acc: 21.13% Original prompt: In the role of a semantic relationship analyzer, examine the connection between the given sentences and decide if they exhibit entailment or not_entailment. Answer with 'entailment' or 'not_entailment': Attacked prompt: b"Around the role of a semantic relationship analyzer, examine the connection between the given sentences and decide if they exhibit entailment or not_entailment. Reactions with 'entailment' or 'not_entailment':" Original acc: 63.38%, attacked acc: 50.70%, dropped acc: 12.68% Original prompt: Working as an entailment classifier, identify whether the given pair of sentences displays entailment or not_entailment. Respond with 'entailment' or 'not_entailment': Attacked prompt: b"Workplace as an entailment classifier, detects whether the given pair of condemning displaying entailment or not_entailment. Answered with 'entailment' or 'not_entailment':" Original acc: 61.97%, attacked acc: 49.30%, dropped acc: 12.68% ## translation Language: chinese, acc: 50.70%, prompt: In the light of an implication analysis tool, evaluate the relationship between the following sentences and classify them as 'entailment' or 'not_entailment'. Language: chinese, acc: 45.07%, prompt: From the perspective of an implication analysis tool, determine whether there is an implication relationship in the following sentences by classifying them as 'entailment' or 'not_entailment'. Language: chinese, acc: 39.44%, prompt: Please use an implication analysis tool to determine whether an implication relationship exists in the following sentences by classifying them as 'entailment' or 'not_entailment'. Language: chinese, acc: 53.52%, prompt: Please evaluate the relation of the following sentences as 'entailment' or 'not_entailment' for the purpose of determining implication relation. Language: chinese, acc: 47.89%, prompt: Please use the implication analysis tool to evaluate the relationships between the following sentences and classify them as 'entailment' or 'not_entailment'. Language: chinese, acc: 39.44%, prompt: For the purpose of determining implicative relations, analyze the relations of the following sentences and classify them as 'entailment' or 'not_entailment'. Language: chinese, acc: 49.30%, prompt: Please use the implication analysis tool to determine the relationship of the following sentences and classify them as 'entailment' or 'not_entailment'. Language: chinese, acc: 39.44%, prompt: Please use the implication judgment tool to assess the relevance of the following sentences and classify them as 'entailment' or 'not_entailment'. Language: chinese, acc: 50.70%, prompt: Please, with implication analysis as the main task, determine the relationships between the following sentences and classify them as 'entailment' or 'not_entailment'. Language: chinese, acc: 38.03%, prompt: Using the implication judgment as a criterion, analyze the relation of the following sentences and classify them as 'entailment' or 'not_entailment'. Language: french, acc: 52.11%, prompt: As an engagement analysis tool, evaluate the relationship between the given sentences and classify it as 'entailment' or 'not_entailment'. Language: french, acc: 52.11%, prompt: Determine whether the given sentences involve one another or not as an implication analysis tool. Classify them accordingly as 'entailment' or 'not_entailment'. Language: french, acc: 54.93%, prompt: Using implication analysis, evaluate whether the sentences provided have a logical relationship and categorize them as 'entailment' or 'not_entailment'. Language: french, acc: 52.11%, prompt: As an engagement assessment tool, determine whether the sentences provided have a logical relationship and classify them as 'entailment' or 'not_entailment'. Language: french, acc: 47.89%, prompt: As an implication classification tool, analyze the sentences provided to determine if there is a logical relationship and categorize them as 'entailment' or 'not_entailment'. Language: french, acc: 26.76%, prompt: Using implication analysis, determine whether the given sentences have a cause-effect relationship and categorize them as 'entailment' or 'not_entailment'. Language: french, acc: 47.89%, prompt: Evaluate the relationship between the given sentences using implication analysis and rank them accordingly as 'entailment' or 'not_entailment'. Language: french, acc: 52.11%, prompt: As an engagement detection tool, determine whether the given sentences have a logical relationship and categorize them as 'entailment' or 'not_entailment'. Language: french, acc: 47.89%, prompt: Using implication analysis, evaluate whether the sentences provided have a cause-effect relationship and rank them accordingly as 'entailment' or 'not_entailment'. Language: french, acc: 36.62%, prompt: Determine whether the given sentences have a cause-effect relationship as an engagement analysis tool and categorize them as 'entailment' or 'not_entailment'. Language: arabic, acc: 56.34%, prompt: In your role as a tool for reasoning analysis, evaluate the relationship between given sentences and classify them as 'entailment' or 'not_entailment'. Language: arabic, acc: 54.93%, prompt: Can you determine whether this sentence is inferred from the other sentence? Classify it as 'entailment' or 'not_entailment'. Language: arabic, acc: 53.52%, prompt: Using the tool of reasoning analysis, analyze the relationship between given sentences and classify them as 'entailment' or 'not_entailment'. Language: arabic, acc: 56.34%, prompt: Does this sentence represent a conclusion from the previous sentence? Classify it as 'entailment' or 'not_entailment'. Language: arabic, acc: 50.70%, prompt: As a tool of reasoning analysis, evaluate the relationship of given sentences and classify them as 'entailment' or 'not_entailment'. Language: arabic, acc: 54.93%, prompt: Can this sentence be inferred from the previous sentence? Classify it as 'entailment' or 'not_entailment'. Language: arabic, acc: 52.11%, prompt: Using a tool to analyze a conclusion, analyze the relationship between the two sentences and classify them as 'entailment' or 'not_entailment'. Language: arabic, acc: 56.34%, prompt: Is this a conclusion from the next sentence? Classify it as 'entailment' or 'not_entailment'. Language: arabic, acc: 52.11%, prompt: As part of your task in analyzing a conclusion, evaluate the relationship between the two sentences and classify them as 'entailment' or 'not_entailment' based on their relationship. Language: arabic, acc: 61.97%, prompt: Are you following this sentence directly from the previous one? Classify it as 'entailment' or 'not_entailment'. Language: spanish, acc: 53.52%, prompt: In your role as an implication analysis tool, evaluate the relationship between the given phrases and classify them as 'entailment' or 'not_entailment'. Language: spanish, acc: 59.15%, prompt: Determine whether the second sentence necessarily implies the first and label the relation as 'entailment', or as 'not_entailment' if not. Language: spanish, acc: 56.34%, prompt: Classifies the relationship between these two sentences as 'entailment' if one necessarily implies the other, or as 'not_entailment' if not. Language: spanish, acc: 59.15%, prompt: Evaluates whether the information in the second sentence is implied in the first and labels the relationship as 'entailment', or as 'not_entailment' if there is no such implication. Language: spanish, acc: 50.70%, prompt: Given a couple of phrases, label their relationship as 'entailment' if one necessarily implies the other, or as 'not_entailment' if there is no such implication. Language: spanish, acc: 52.11%, prompt: Analyzes the relationship between the phrases and classifies them as 'entailment' if one necessarily implies the other, or as 'not_entailment' if there is no such implication. Language: spanish, acc: 56.34%, prompt: Given two sentences, determine whether the second sentence is a necessary consequence of the first and label the relation as 'entailment', or as 'not_entailment' if not. Language: spanish, acc: 56.34%, prompt: Evaluates whether the information presented in the second sentence is implicit in the first and labels the relationship as 'entailment', or as 'not_entailment' if there is no such implication. Language: spanish, acc: 54.93%, prompt: Classifies the relationship between the given phrases as 'entailment' if one necessarily implies the other, or as 'not_entailment' if there is no such implication. Language: spanish, acc: 59.15%, prompt: Determines whether the information provided in the second sentence is necessarily inferable from the first and labels the relationship as 'entailment', or as 'not_entailment' if not. Language: japanese, acc: 53.52%, prompt: Analyze the relationship of a given sentence and classify it as 'entailment' or 'not_entailment'. Language: japanese, acc: 54.93%, prompt: Evaluate the semantic relationship of the sentence and classify it as 'entailment' or 'not_entailment'. Language: japanese, acc: 52.11%, prompt: Please judge the relationship between the given sentences and classify them as 'entailment' or 'not_entailment'. Language: japanese, acc: 50.70%, prompt: Examine the paraphrases of a given sentence and classify them 'entailment' or 'not_entailment'. Language: japanese, acc: 45.07%, prompt: Rate the similarity of a given sentence and categorize it as 'entailment' or 'not_entailment'. Language: japanese, acc: 47.89%, prompt: Determinate the semantic connections of a given sentence and classify it as 'entailment' or 'not_entailment'. Language: japanese, acc: 53.52%, prompt: Examine the semantic match of a given sentence and categorize it as 'entailment' or 'not_entailment'. Language: japanese, acc: 53.52%, prompt: Classify it as 'entailment' or 'not_entailment' based on the content of the sentence. Language: japanese, acc: 53.52%, prompt: Analyze the relationship of a given sentence and classify it as 'entailment' or 'not_entailment'. Language: japanese, acc: 50.70%, prompt: Judge the semantic connections of a given sentence and categorize it as 'entailment' or 'not_entailment'. Language: korean, acc: 49.30%, prompt: Evaluate the relationship between any two sentences given to you and classify you as 'entailment' or 'not_entailment'. Language: korean, acc: 52.11%, prompt: Analyze the semantic deductive relations between sentences and classify them as 'entailment' or 'not_entailment'. Language: korean, acc: 50.70%, prompt: Evaluate the logical relevance between sentences and classify them as 'entailment' or 'not_entailment'. Language: korean, acc: 52.11%, prompt: Evaluate the interaction of two given sentences and classify them as 'entailment' or 'not_entailment'. Language: korean, acc: 61.97%, prompt: Please check whether there is a semantic match between those two sentences and classify them as 'entailment' or 'not_entailment'. Language: korean, acc: 52.11%, prompt: Compare information between given sentences and classify them as 'entailment' or 'not_entailment'. Language: korean, acc: 46.48%, prompt: Please analyse the correlation between those two sentences and classify them as 'entailment' or 'not_entailment'. Language: korean, acc: 53.52%, prompt: Evaluate the different meanings between given sentences and classify them as 'entailment' or 'not_entailment'. Language: korean, acc: 42.25%, prompt: Compare the semantic structure of the two sentences and classify them as 'entailment' or 'not_entailment'. Language: korean, acc: 49.30%, prompt: Evaluate the interactions between sentences and classify them as 'entailment' or 'not_entailment'. # mmlu ## 10 prompts Acc: 64.00%, prompt: In relation to the multiple-choice question on {}, please provide the accurate answer by choosing 'A', 'B', 'C', or 'D'. Acc: 63.50%, prompt: Regarding the following multiple-choice question on {}, pick the correct answer from the options 'A', 'B', 'C', or 'D'. Acc: 62.50%, prompt: For each multiple-choice question about {}, identify the correct answer by selecting 'A', 'B', 'C', or 'D'. Acc: 62.00%, prompt: For the multiple-choice question related to {}, please choose the most accurate answer from 'A', 'B', 'C', or 'D'. Acc: 61.50%, prompt: Answer the following multiple-choice question about {} by selecting the correct option: 'A', 'B', 'C', or 'D'. Acc: 61.50%, prompt: Examine the following question based on {} and choose the correct response from 'A', 'B', 'C', or 'D'. Acc: 61.00%, prompt: Below are multiple-choice question concerning {}. Indicate your response with 'A', 'B', 'C', or 'D'. Acc: 61.00%, prompt: Please respond to the multiple-choice question about {} by selecting the appropriate answer: 'A', 'B', 'C', or 'D'. Acc: 61.00%, prompt: Answer the subsequent multiple-choice question about {} by picking the right option among 'A', 'B', 'C', or 'D'. Acc: 57.00%, prompt: Evaluate the multiple-choice question about {} and select the most fitting response from 'A', 'B', 'C', or 'D'. Acc: 62.50%, prompt: As a skilled person in the field of {}, please respond to the multiple-choice question by choosing 'A', 'B', 'C', or 'D'. Acc: 62.00%, prompt: With your knowledge of {}, tackle the following multiple-choice question by choosing 'A', 'B', 'C', or 'D'. Acc: 62.00%, prompt: Drawing upon your mastery of {}, please answer the multiple-choice question by selecting the correct option from 'A', 'B', 'C', or 'D'. Acc: 61.50%, prompt: As an expert in {}, respond to the following multiple-choice question by selecting 'A', 'B', 'C', or 'D'. Acc: 61.50%, prompt: With your understanding of {}, kindly answer the subsequent multiple-choice question by selecting 'A', 'B', 'C', or 'D'. Acc: 60.00%, prompt: Utilizing your expertise in {}, answer the following multiple-choice question by picking 'A', 'B', 'C', or 'D'. Acc: 60.00%, prompt: As a knowledgeable individual in {}, provide your response to the multiple-choice question by choosing 'A', 'B', 'C', or 'D'. Acc: 59.50%, prompt: As someone well-versed in {}, please address the multiple-choice question below by selecting 'A', 'B', 'C', or 'D'. Acc: 59.50%, prompt: Considering your familiarity with {}, attend to the following multiple-choice question by picking 'A', 'B', 'C', or 'D'. Acc: 59.00%, prompt: Given your proficiency in {}, please answer the subsequent multiple-choice question with 'A', 'B', 'C', or 'D'. ## bertattack Original prompt: Regarding the following multiple-choice question on {}, pick the correct answer from the options 'A', 'B', 'C', or 'D'. Attacked prompt: b"1 the first multiple-choice question on {}, remove the following answer from the options 'A', 'B', 'C', or 'D'. " Original acc: 63.50%, attacked acc: 14.50%, dropped acc: 49.00% Original prompt: In relation to the multiple-choice question on {}, please provide the accurate answer by choosing 'A', 'B', 'C', or 'D'. Attacked prompt: b"In regard to the multiple-choice question on {}, two express the accurate conclusion by choosing 'A', 'B', 'C', or 'D'. " Original acc: 64.00%, attacked acc: 16.50%, dropped acc: 47.50% Original prompt: For each multiple-choice question about {}, identify the correct answer by selecting 'A', 'B', 'C', or 'D'. Attacked prompt: b"after each multiple-response question about {}, suggest the exact subject by selecting 'A', 'B', 'C', or 'D'. " Original acc: 63.00%, attacked acc: 29.00%, dropped acc: 34.00% Original prompt: As a skilled person in the field of {}, please respond to the multiple-choice question by choosing 'A', 'B', 'C', or 'D'. Attacked prompt: b"1 a possible person in the field of {}, please question to the multiple-selection question by choosing 'A', 'B', 'C', or 'D'." Original acc: 62.00%, attacked acc: 52.50%, dropped acc: 9.50% Original prompt: With your knowledge of {}, tackle the following multiple-choice question by choosing 'A', 'B', 'C', or 'D'. Attacked prompt: b"With your grasp of {}, question the using multiple-response question by choosing 'A', 'B', 'C', or 'D'." Original acc: 61.50%, attacked acc: 27.00%, dropped acc: 34.50% Original prompt: Drawing upon your mastery of {}, please answer the multiple-choice question by selecting the correct option from 'A', 'B', 'C', or 'D'. Attacked prompt: b"Drawing upon your mastery of {}, simply question the multiple-selecting question by selecting the wrong color from 'A', 'B', 'C', or 'D'." Original acc: 62.00%, attacked acc: 1.00%, dropped acc: 61.00% ## checklist Original prompt: In relation to the multiple-choice question on {}, please provide the accurate answer by choosing 'A', 'B', 'C', or 'D'. Attacked prompt: b"In relation to the multiple-choice question on {}, please provide the accurate answer by choosing 'A', 'B', 'C', or 'D' 5yWbBXztUY. " Original acc: 64.50%, attacked acc: 59.50%, dropped acc: 5.00% Original prompt: Regarding the following multiple-choice question on {}, pick the correct answer from the options 'A', 'B', 'C', or 'D'. Attacked prompt: b"Regarding the following multiple-choice question on {}, pick the correct answer from the options 'A', 'B', 'C', or 'D' Pbl3tYuoRo. " Original acc: 64.00%, attacked acc: 58.00%, dropped acc: 6.00% Original prompt: For each multiple-choice question about {}, identify the correct answer by selecting 'A', 'B', 'C', or 'D'. Attacked prompt: b"For each multiple-choice question about {}, identify the correct answer by selecting 'A', 'B', 'C', or 'D' d6ZQ3u0GBQ. " Original acc: 63.00%, attacked acc: 58.00%, dropped acc: 5.00% Original prompt: With your knowledge of {}, tackle the following multiple-choice question by choosing 'A', 'B', 'C', or 'D'. Attacked prompt: b"With your knowledge of {}, tackle the following multiple-choice question by choosing 'A', 'B', 'C', or 'D' 5yWbBXztUY." Original acc: 62.50%, attacked acc: 59.50%, dropped acc: 3.00% Original prompt: As a skilled person in the field of {}, please respond to the multiple-choice question by choosing 'A', 'B', 'C', or 'D'. Attacked prompt: b"As a skilled person in the field of {}, please respond to the multiple-choice question by choosing 'A', 'B', 'C', or 'D' IhuBIhoPGc." Original acc: 61.50%, attacked acc: 58.00%, dropped acc: 3.50% Original prompt: Drawing upon your mastery of {}, please answer the multiple-choice question by selecting the correct option from 'A', 'B', 'C', or 'D'. Attacked prompt: b"Drawing upon your mastery of {}, please answer the multiple-choice question by selecting the correct option from 'A', 'B', 'C', or 'D' IhuBIhoPGc." Original acc: 61.50%, attacked acc: 59.50%, dropped acc: 2.00% ## deepwordbug Original prompt: In relation to the multiple-choice question on {}, please provide the accurate answer by choosing 'A', 'B', 'C', or 'D'. Attacked prompt: b"nI relatFon to the multiple-choice question on {}, please provcide the accurate answer by choosing 'A', 'B', 'C', or 'D'. " Original acc: 64.50%, attacked acc: 60.50%, dropped acc: 4.00% Original prompt: Regarding the following multiple-choice question on {}, pick the correct answer from the options 'A', 'B', 'C', or 'D'. Attacked prompt: b"Regarding the following multiple-choice question on {}, pick the correct answeWr from the options 'A', 'B', 'C', or 'D'. " Original acc: 62.50%, attacked acc: 60.00%, dropped acc: 2.50% Original prompt: For each multiple-choice question about {}, identify the correct answer by selecting 'A', 'B', 'C', or 'D'. Attacked prompt: b"For each multiple-choice question about {}, identUfy the corect ansewer by selecting 'A', 'B', 'C', or 'D'. " Original acc: 63.00%, attacked acc: 60.00%, dropped acc: 3.00% Original prompt: As a skilled person in the field of {}, please respond to the multiple-choice question by choosing 'A', 'B', 'C', or 'D'. Attacked prompt: b"As a skilled person in the field of {}, please respond to the multiple-choice question by choosing 'A', 'B', 'C', or 'D'." Original acc: 61.50%, attacked acc: 61.50%, dropped acc: 1.50% Original prompt: With your knowledge of {}, tackle the following multiple-choice question by choosing 'A', 'B', 'C', or 'D'. Attacked prompt: b"With your knowledge of {}, tackle the follQwing multiple-choice question by choosing 'A', 'B', 'C', or 'D'." Original acc: 62.50%, attacked acc: 59.00%, dropped acc: 3.50% Original prompt: Drawing upon your mastery of {}, please answer the multiple-choice question by selecting the correct option from 'A', 'B', 'C', or 'D'. Attacked prompt: b"Drawing upon your mastery of {}, please anwser the multiple-choice question by selecting the correct option from 'A', 'B', 'C', or 'D'." Original acc: 62.00%, attacked acc: 59.00%, dropped acc: 3.00% ## stresstest Original prompt: In relation to the multiple-choice question on {}, please provide the accurate answer by choosing 'A', 'B', 'C', or 'D'. Attacked prompt: b"In relation to the multiple-choice question on {}, please provide the accurate answer by choosing 'A', 'B', 'C', or 'D' and true is true . " Original acc: 65.00%, attacked acc: 60.00%, dropped acc: 5.00% Original prompt: Regarding the following multiple-choice question on {}, pick the correct answer from the options 'A', 'B', 'C', or 'D'. Attacked prompt: b"Regarding the following multiple-choice question on {}, pick the correct answer from the options 'A', 'B', 'C', or 'D' and true is true . " Original acc: 63.00%, attacked acc: 56.50%, dropped acc: 6.50% Original prompt: For each multiple-choice question about {}, identify the correct answer by selecting 'A', 'B', 'C', or 'D'. Attacked prompt: b"For each multiple-choice question about {}, identify the correct answer by selecting 'A', 'B', 'C', or 'D' and true is true . " Original acc: 63.00%, attacked acc: 53.50%, dropped acc: 9.50% Original prompt: With your knowledge of {}, tackle the following multiple-choice question by choosing 'A', 'B', 'C', or 'D'. Attacked prompt: b"With your knowledge of {}, tackle the following multiple-choice question by choosing 'A', 'B', 'C', or 'D' and false is not true ." Original acc: 62.00%, attacked acc: 58.00%, dropped acc: 4.00% Original prompt: As a skilled person in the field of {}, please respond to the multiple-choice question by choosing 'A', 'B', 'C', or 'D'. Attacked prompt: b"As a skilled person in the field of {}, please respond to the multiple-choice question by choosing 'A', 'B', 'C', or 'D' and false is not true ." Original acc: 61.50%, attacked acc: 57.00%, dropped acc: 4.50% Original prompt: Drawing upon your mastery of {}, please answer the multiple-choice question by selecting the correct option from 'A', 'B', 'C', or 'D'. Attacked prompt: b"Drawing upon your mastery of {}, please answer the multiple-choice question by selecting the correct option from 'A', 'B', 'C', or 'D' and true is true ." Original acc: 61.00%, attacked acc: 57.00%, dropped acc: 4.00% ## textbugger Original prompt: In relation to the multiple-choice question on {}, please provide the accurate answer by choosing 'A', 'B', 'C', or 'D'. Attacked prompt: b"In relation to the multiple-choice issue on {}, ple\xc9\x91se delivering the exact answers by choosing 'A', 'B', 'C', or 'D'. " Original acc: 64.50%, attacked acc: 58.50%, dropped acc: 6.00% Original prompt: Regarding the following multiple-choice question on {}, pick the correct answer from the options 'A', 'B', 'C', or 'D'. Attacked prompt: b"Regarding the hereafter multiple-choice question on {}, pi\xcf\xb2k the corrects answers from the options 'A', 'B', 'C', or 'D'. " Original acc: 62.50%, attacked acc: 51.00%, dropped acc: 11.50% Original prompt: For each multiple-choice question about {}, identify the correct answer by selecting 'A', 'B', 'C', or 'D'. Attacked prompt: b"For each multipl\xd0\xb5-choice question about {}, identify the correct answers by \xd1\x95electing 'A', 'B', 'C', or 'D'. " Original acc: 63.00%, attacked acc: 56.00%, dropped acc: 7.00% Original prompt: With your knowledge of {}, tackle the following multiple-choice question by choosing 'A', 'B', 'C', or 'D'. Attacked prompt: b"Wit\xd5\xb0 your knowledge of {}, confront the following multip\xe2\x85\xbce-choice question by select 'A', 'B', 'C', or 'D'." Original acc: 62.50%, attacked acc: 57.50%, dropped acc: 5.00% Original prompt: As a skilled person in the field of {}, please respond to the multiple-choice question by choosing 'A', 'B', 'C', or 'D'. Attacked prompt: b"Since a skill\xd0\xb5d person in the \xf0\x9d\x9a\x8field of {}, please respond to the multiple-choice question by choosing 'A', 'B', 'C', or 'D'." Original acc: 62.50%, attacked acc: 58.50%, dropped acc: 4.00% Original prompt: Drawing upon your mastery of {}, please answer the multiple-choice question by selecting the correct option from 'A', 'B', 'C', or 'D'. Attacked prompt: b"Drawing upon your mastery of {}, please answer the multiple-choice question by selecting the corre\xcf\xb2t option from 'A', 'B', 'C', or 'D'." Original acc: 61.50%, attacked acc: 59.00%, dropped acc: 2.50% ## textfooler Original prompt: Regarding the following multiple-choice question on {}, pick the correct answer from the options 'A', 'B', 'C', or 'D'. Attacked prompt: b"Regarding the henceforth multiple-choice subjects on {}, select the correct rebuttal from the options 'A', 'B', 'C', or 'D'. " Original acc: 63.50%, attacked acc: 52.00%, dropped acc: 11.50% Original prompt: In relation to the multiple-choice question on {}, please provide the accurate answer by choosing 'A', 'B', 'C', or 'D'. Attacked prompt: b"In relation to the multiple-choice issue on {}, please delivering the clarification rebuttal by selector 'A', 'B', 'C', or 'D'. " Original acc: 64.00%, attacked acc: 50.50%, dropped acc: 13.50% Original prompt: For each multiple-choice question about {}, identify the correct answer by selecting 'A', 'B', 'C', or 'D'. Attacked prompt: b"For each multiple-choice question about {}, discovered the correct rebuttal by picks 'A', 'B', 'C', or 'D'. " Original acc: 62.50%, attacked acc: 53.50%, dropped acc: 9.00% Original prompt: With your knowledge of {}, tackle the following multiple-choice question by choosing 'A', 'B', 'C', or 'D'. Attacked prompt: b"Among your knowledge of {}, cope the following multiple-choice question by choosing 'A', 'B', 'C', or 'D'." Original acc: 62.00%, attacked acc: 58.50%, dropped acc: 3.50% Original prompt: As a skilled person in the field of {}, please respond to the multiple-choice question by choosing 'A', 'B', 'C', or 'D'. Attacked prompt: b"As a expert person in the minefields of {}, please reacts to the multiple-choice question by choice 'A', 'B', 'C', or 'D'." Original acc: 61.50%, attacked acc: 54.50%, dropped acc: 7.00% Original prompt: Drawing upon your mastery of {}, please answer the multiple-choice question by selecting the correct option from 'A', 'B', 'C', or 'D'. Attacked prompt: b"Drawing upon your masters of {}, please answer the multiple-choice question by selecting the exact opting from 'A', 'B', 'C', or 'D'." Original acc: 61.50%, attacked acc: 58.50%, dropped acc: 3.00% ## translation Language: chinese, acc: 64.00%, prompt: For the following multiple choice question about {}, please select the correct answer from 'A', 'B', 'C' or 'D'. Language: chinese, acc: 62.50%, prompt: Please select the correct answer from 'A', 'B', 'C' or 'D' for the following multiple choice question for {}. Language: chinese, acc: 62.00%, prompt: Select the correct answer from 'A', 'B', 'C' or 'D' for the following multiple choice question {}. Language: chinese, acc: 64.00%, prompt: Please choose the correct answer from 'A', 'B', 'C' or 'D' according to the following multiple-choice questions about {}. Language: chinese, acc: 62.50%, prompt: Please select the correct answer from 'A', 'B', 'C' or 'D' for the {} multiple choice questions below. Language: chinese, acc: 62.00%, prompt: The following is A multiple choice question about {}. Please select the correct answer from 'A', 'B', 'C' or 'D'. Language: chinese, acc: 64.50%, prompt: Please select the correct answer from 'A', 'B', 'C' or 'D' for the following multiple choice question {}. Language: chinese, acc: 63.00%, prompt: Please choose the correct answer from 'A', 'B', 'C' or 'D' according to the following multiple-choice questions about {}. Language: chinese, acc: 63.00%, prompt: Please select the correct answer from 'A', 'B', 'C' or 'D' for the following multiple choice questions about {}. Language: chinese, acc: 62.50%, prompt: Please select the correct answer from 'A', 'B', 'C' or 'D' for the following multiple choice questions about {}. Language: french, acc: 65.00%, prompt: For the following multiple choice question on {}, choose the correct answer from options 'A', 'B', 'C' or 'D'. Language: french, acc: 62.50%, prompt: This is a multiple choice question about {}. Select the correct answer from options 'A', 'B', 'C' or 'D'. Language: french, acc: 61.00%, prompt: In the context of the multiple-choice question on {}, identify the correct answer from options 'A', 'B', 'C' or 'D'. Language: french, acc: 61.50%, prompt: About the following question on {}, determine the correct answer from the choices 'A', 'B', 'C' or 'D'. Language: french, acc: 61.50%, prompt: Carefully review the multiple-choice question regarding {}. Choose the correct answer from options 'A', 'B', 'C', or 'D'. Language: french, acc: 62.50%, prompt: For the multiple-choice question for {}, indicate the correct answer from options 'A', 'B', 'C', or 'D'. Language: french, acc: 64.00%, prompt: The next question is about {}. Select the correct answer from the choices 'A', 'B', 'C' or 'D'. Language: french, acc: 63.50%, prompt: As part of the multiple-choice question on {}, choose the appropriate answer from options 'A', 'B', 'C' or 'D'. Language: french, acc: 52.00%, prompt: Rate your understanding of the multiple-choice question on {}. Choose the correct answer from options 'A', 'B', 'C' or 'D'. Language: french, acc: 59.50%, prompt: Analyze the following multiple-choice question on {}. Identify the correct answer among choices 'A', 'B', 'C' or 'D'. Language: arabic, acc: 63.50%, prompt: For the multiple choice question about {}, choose the correct answer from options 'A', 'B', 'C' or 'D'. Language: arabic, acc: 63.50%, prompt: For the following multiple-choice question about {}, choose the correct answer from options 'A', 'B', 'C' or 'D'. Language: arabic, acc: 63.00%, prompt: For the following multiple choice question about {}, choose the correct answer from options 'A', 'B', 'C' or 'D'. Language: arabic, acc: 62.00%, prompt: When it comes to the multiple-choice question about {}, choose the correct answer from options 'A', 'B', 'C' or 'D'. Language: arabic, acc: 62.00%, prompt: For the multiple-choice question about {}, choose the correct answer from options 'A', 'B', 'C' or 'D'. Language: arabic, acc: 62.00%, prompt: If the question for {} is multiple choice, choose the correct answer from options 'A', 'B', 'C' or 'D'. Language: arabic, acc: 60.50%, prompt: For the question regarding {}, choose the correct answer from options 'A', 'B', 'C' or 'D'. Language: arabic, acc: 61.00%, prompt: For the question about {}, choose the correct answer from options 'A', 'B', 'C' or 'D'. Language: arabic, acc: 61.00%, prompt: When it comes to the question regarding {}, choose the correct answer from options 'A', 'B', 'C' or 'D'. Language: arabic, acc: 61.00%, prompt: For the question regarding {}, choose the correct answer from options 'A', 'B', 'C' or 'D'. Language: spanish, acc: 64.00%, prompt: For the following multiple-choice question about {}, choose the correct answer from 'A', 'B', 'C', or 'D'. Language: spanish, acc: 64.50%, prompt: For the following multiple-choice question about {}, select the correct answer from 'A', 'B', 'C', or 'D'. Language: spanish, acc: 63.50%, prompt: For the following multiple-choice question about {}, choose the correct answer from 'A', 'B', 'C', or 'D'. Language: spanish, acc: 61.00%, prompt: Within the context of the following multiple-choice question about {}, choose the correct option from 'A', 'B', 'C', or 'D'. Language: spanish, acc: 62.00%, prompt: For the following multiple-choice statement about {}, select the correct answer from 'A', 'B', 'C', or 'D'. Language: spanish, acc: 60.50%, prompt: Considering the following multiple-choice question about {}, mark the correct answer with 'A', 'B', 'C', or 'D'. Language: spanish, acc: 64.00%, prompt: For the following multiple-choice question about {}, choose the correct alternative among 'A', 'B', 'C' or 'D'. Language: spanish, acc: 62.50%, prompt: For the following multiple-choice statement about {}, choose the correct option from alternatives 'A', 'B', 'C', or 'D'. Language: spanish, acc: 61.00%, prompt: Within the context of the following multiple-choice question about {}, select the correct answer from alternatives 'A', 'B', 'C', or 'D'. Language: spanish, acc: 60.00%, prompt: Considering the following multiple-choice statement about {}, mark the correct alternative with the options 'A', 'B', 'C' or 'D'. Language: japanese, acc: 65.50%, prompt: Choose the appropriate answer from options 'A', 'B', 'C', or 'D' for {} regarding the following question. Language: japanese, acc: 63.50%, prompt: Choose the correct answer from 'A', 'B', 'C', or 'D' for the following multiple-choice question about {}. Language: japanese, acc: 64.50%, prompt: For the following multiple-choice questions about {}, choose the correct answer from 'A', 'B', 'C', or 'D'. Language: japanese, acc: 62.50%, prompt: Choose the correct answer from options 'A', 'B', 'C', or 'D' for the following questions about {}. Language: japanese, acc: 61.50%, prompt: In the multiple choice questions below, choose the correct answer for {} from 'A', 'B', 'C', or 'D'. Language: japanese, acc: 63.00%, prompt: Choose the correct answer from the options 'A', 'B', 'C', or 'D' for the following questions about {}. Language: japanese, acc: 61.50%, prompt: In the multiple choice questions below, choose the correct answer for {} from 'A', 'B', 'C', or 'D'. Language: japanese, acc: 63.00%, prompt: Choose the correct answer from 'A', 'B', 'C', or 'D' for the following multiple choice questions about {}. Language: japanese, acc: 61.00%, prompt: In the multiple choice questions below, choose the correct answer for {} from 'A', 'B', 'C', or 'D'. Language: japanese, acc: 64.00%, prompt: Choose the correct answer from options 'A', 'B', 'C', or 'D' for {} regarding the following question. Language: korean, acc: 53.00%, prompt: For the multiple choice problem about, choose the correct answer for '{}' from 'A', 'B', 'C', or 'D'. Language: korean, acc: 56.50%, prompt: Choose the correct answer for '{}' from 'A', 'B', 'C', or 'D' in the multiple choice problem involving, Language: korean, acc: 54.50%, prompt: For the multiple choice problem below, choose the correct answer to '{}' from 'A', 'B', 'C', or 'D'. Language: korean, acc: 54.50%, prompt: In the following multiple-choice problem, choose the correct answer for '{}' from 'A', 'B', 'C', or 'D'. Language: korean, acc: 54.00%, prompt: For the following multiple choice problem, choose the correct answer for '{}' from 'A', 'B', 'C', or 'D'. Language: korean, acc: 58.00%, prompt: Solve multiple choice problems about: Which of 'A', 'B', 'C', or 'D' is the correct answer for '{}'. Language: korean, acc: 49.50%, prompt: Choose the correct answer to the multiple-choice question below. Is '{}' an 'A', 'B', 'C', or 'D'. Language: korean, acc: 60.50%, prompt: Solve the following multiple-choice problem. Choose the correct answer for '{}' from 'A', 'B', 'C', or 'D'. Language: korean, acc: 53.00%, prompt: Choose the correct answer to the following multiple choice problem: Is '{}' 'A', 'B', 'C', or 'D'. Language: korean, acc: 60.50%, prompt: Solve multiple-choice problems about: Please select 'A', 'B', 'C', or 'D' for the correct answer to '{}'. # squad_v2 ## 10 prompts Acc: 72.37%, prompt: Discover the best answer based on the context. If the context doesn't include an answer, respond with 'unanswerable'. Acc: 70.16%, prompt: Using the context, determine the most suitable answer. If the context doesn't contain the answer, respond with 'unanswerable'. Acc: 69.99%, prompt: Based on the given context, provide the best possible answer. If there's no answer available in the context, respond with 'unanswerable'. Acc: 68.98%, prompt: Please derive the most fitting answer from the context. If there isn't an answer in the context, respond with 'unanswerable'. Acc: 67.97%, prompt: Locate the most accurate answer within the context. If the context doesn't provide an answer, respond with 'unanswerable'. Acc: 67.89%, prompt: Find the correct answer in the context provided. If an answer cannot be found, please respond with 'unanswerable'. Acc: 66.29%, prompt: From the context, provide the most precise answer. If the answer is not in the context, respond with 'unanswerable'. Acc: 63.69%, prompt: Please extract the most appropriate answer from the context. If an answer is not present, indicate 'unanswerable'. Acc: 61.72%, prompt: Search the context for the most relevant answer. If the answer cannot be found, respond with 'unanswerable'. Acc: 60.58%, prompt: Identify the most relevant answer from the context. If it's not possible to find an answer, respond with 'unanswerable'. Acc: 68.93%, prompt: Using your knowledge of the context, identify the best answer to the question. If the context doesn't provide an answer, write 'unanswerable'. Acc: 65.09%, prompt: As an expert with a deep understanding of the context, find the best answer. If the context doesn't include an answer, say 'unanswerable'. Acc: 64.97%, prompt: Drawing on your expertise in the context, provide the most precise answer. If the answer is not in the context, respond with 'unanswerable'. Acc: 64.81%, prompt: With your extensive knowledge of the context, answer the question accurately. If the context doesn't contain the answer, reply with 'unanswerable'. Acc: 64.40%, prompt: Being well-versed in the context, please derive the most fitting answer. If there isn't an answer in the context, use 'unanswerable'. Acc: 63.21%, prompt: Drawing upon your expertise in the context, determine the most suitable answer. If an answer isn't available, state 'unanswerable'. Acc: 62.88%, prompt: As a subject matter expert, extract the correct answer from the context. If an answer is not present, indicate 'unanswerable'. Acc: 61.22%, prompt: As a proficient expert in the given context, search for the most relevant answer. If the answer cannot be found, respond by saying 'unanswerable'. Acc: 58.74%, prompt: As an authority on the context, locate the most accurate answer. If the context doesn't contain the answer, mention 'unanswerable'. Acc: 57.90%, prompt: As a well-informed specialist familiar with the context, provide an answer to the question. If the context doesn't contain an answer, reply with 'unanswerable'. ## bertattack Original prompt: Discover the best answer based on the context. If the context doesn't include an answer, respond with 'unanswerable'. Attacked prompt: b"challenge the best answer based on the context. however the context doesn't include an answer, play with 'unanswerable'." Original acc: 72.47%, attacked acc: 2.76%, dropped acc: 69.71% Original prompt: Using the context, determine the most suitable answer. If the context doesn't contain the answer, respond with 'unanswerable'. Attacked prompt: b"choosing the context, determine the most suitable answer. pretend the context doesn't contain the outcome, construct with 'unanswerable'." Original acc: 71.16%, attacked acc: 24.67%, dropped acc: 46.49% Original prompt: Based on the given context, provide the best possible answer. If there's no answer available in the context, respond with 'unanswerable'. Attacked prompt: b"by on the given context, provide the perfect possible answer. pretending there'tis no answer but in the context, play with 'unanswerable'." Original acc: 69.60%, attacked acc: 5.53%, dropped acc: 64.08% Original prompt: Using your knowledge of the context, identify the best answer to the question. If the context doesn't provide an answer, write 'unanswerable'. Attacked prompt: b"may your understanding of the context, identify the best answer to the question. while the text doesn't provided an answer, search 'unanswerable'." Original acc: 69.21%, attacked acc: 27.18%, dropped acc: 42.03% Original prompt: As an expert with a deep understanding of the context, find the best answer. If the context doesn't include an answer, say 'unanswerable'. Attacked prompt: b"ask an expertise with a deep understands of the context, create the most explanation. whether the context doesn't warrant an answer, look 'unanswerable'." Original acc: 65.60%, attacked acc: 13.02%, dropped acc: 52.58% Original prompt: Drawing on your expertise in the context, provide the most precise answer. If the answer is not in the context, respond with 'unanswerable'. Attacked prompt: b"Drawing on your technique in the frame, offer the most precise answer. saying the answer is not in the context, play with 'unanswerable'." Original acc: 65.63%, attacked acc: 1.86%, dropped acc: 63.77% ## checklist Original prompt: Discover the best answer based on the context. If the context doesn't include an answer, respond with 'unanswerable'. Attacked prompt: b"Discover the best answer based on the context. If the context doesn't include an answer, respond with 'unanswerable' 5yWbBXztUY." Original acc: 72.37%, attacked acc: 68.11%, dropped acc: 4.26% Original prompt: Using the context, determine the most suitable answer. If the context doesn't contain the answer, respond with 'unanswerable'. Attacked prompt: b"Using the context, determine the most suitable answer. If the context doesn't contain the answer, respond with 'unanswerable' ItGDrrA1Sc." Original acc: 71.41%, attacked acc: 68.81%, dropped acc: 2.60% Original prompt: Based on the given context, provide the best possible answer. If there's no answer available in the context, respond with 'unanswerable'. Attacked prompt: b"Based on the given context, provide the best possible answer. If there's no answer available in the context, respond with 'unanswerable' ItGDrrA1Sc." Original acc: 69.16%, attacked acc: 64.99%, dropped acc: 4.17% Original prompt: Using your knowledge of the context, identify the best answer to the question. If the context doesn't provide an answer, write 'unanswerable'. Attacked prompt: b"Using your knowledge of the context, identify the best answer to the question. If the context doesn't provide an answer, write 'unanswerable' lWJoGGRdjv." Original acc: 68.89%, attacked acc: 68.47%, dropped acc: 0.42% Original prompt: Drawing on your expertise in the context, provide the most precise answer. If the answer is not in the context, respond with 'unanswerable'. Attacked prompt: b"Drawing on your expertise in the context, provide the most precise answer. If the answer is not in the context, respond with 'unanswerable' fwhUZUQzXW." Original acc: 65.30%, attacked acc: 60.83%, dropped acc: 4.48% Original prompt: As an expert with a deep understanding of the context, find the best answer. If the context doesn't include an answer, say 'unanswerable'. Attacked prompt: b"As an expert with a deep understanding of the context, find the best answer. If the context doesn't include an answer, say 'unanswerable' d6ZQ3u0GBQ." Original acc: 64.31%, attacked acc: 60.20%, dropped acc: 4.11% ## deepwordbug Original prompt: Discover the best answer based on the context. If the context doesn't include an answer, respond with 'unanswerable'. Attacked prompt: b"DiscoveMr the best answr bassd on the Rontext. Iof the cnotext doesn't include an answer, resCpond with 'unanswerable'." Original acc: 71.87%, attacked acc: 56.61%, dropped acc: 15.26% Original prompt: Using the context, determine the most suitable answer. If the context doesn't contain the answer, respond with 'unanswerable'. Attacked prompt: b"Using the cotnext, determine the most suitbale answer. If the context doesn't contain the answer, respond with 'unanswerable'." Original acc: 70.41%, attacked acc: 65.68%, dropped acc: 4.73% Original prompt: Based on the given context, provide the best possible answer. If there's no answer available in the context, respond with 'unanswerable'. Attacked prompt: b"Based on the giFven conteMt, pKrovide the besu possible banswer. If ther'es no answer avarlable in the contxt, resCpond with 'unanswerable'." Original acc: 70.01%, attacked acc: 58.03%, dropped acc: 11.98% Original prompt: Using your knowledge of the context, identify the best answer to the question. If the context doesn't provide an answer, write 'unanswerable'. Attacked prompt: b"Usign your knowledge of the ontext, identifBy the best answer to the question. If the contWxt doesn't pEovide an ansOer, wrLte 'unanswerable'." Original acc: 69.24%, attacked acc: 59.87%, dropped acc: 9.37% Original prompt: Drawing on your expertise in the context, provide the most precise answer. If the answer is not in the context, respond with 'unanswerable'. Attacked prompt: b"rDawing on your expertise in the conQext, provide the most rpecise answer. If the answer is not in the contexW, respond with 'unanswerable'." Original acc: 64.79%, attacked acc: 54.06%, dropped acc: 10.74% Original prompt: As an expert with a deep understanding of the context, find the best answer. If the context doesn't include an answer, say 'unanswerable'. Attacked prompt: b"zs an exprt with a deep understanding of the context, find the best answr. If the context doesn't include an anCswer, asy 'unanswerable'." Original acc: 64.50%, attacked acc: 36.52%, dropped acc: 27.97% ## stresstest Original prompt: Discover the best answer based on the context. If the context doesn't include an answer, respond with 'unanswerable'. Attacked prompt: b"Discover the best answer based on the context. If the context doesn't include an answer, respond with 'unanswerable' and true is true and true is true and true is true and true is true and true is true ." Original acc: 72.37%, attacked acc: 71.94%, dropped acc: 0.43% Original prompt: Using the context, determine the most suitable answer. If the context doesn't contain the answer, respond with 'unanswerable'. Attacked prompt: b"Using the context, determine the most suitable answer. If the context doesn't contain the answer, respond with 'unanswerable' and true is true and true is true and true is true and true is true and true is true ." Original acc: 70.66%, attacked acc: 71.10%, dropped acc: -0.44% Original prompt: Based on the given context, provide the best possible answer. If there's no answer available in the context, respond with 'unanswerable'. Attacked prompt: b"Based on the given context, provide the best possible answer. If there's no answer available in the context, respond with 'unanswerable' and true is true ." Original acc: 69.30%, attacked acc: 69.88%, dropped acc: -0.57% Original prompt: Using your knowledge of the context, identify the best answer to the question. If the context doesn't provide an answer, write 'unanswerable'. Attacked prompt: b"Using your knowledge of the context, identify the best answer to the question. If the context doesn't provide an answer, write 'unanswerable' and true is true ." Original acc: 69.80%, attacked acc: 69.83%, dropped acc: -0.03% Original prompt: Drawing on your expertise in the context, provide the most precise answer. If the answer is not in the context, respond with 'unanswerable'. Attacked prompt: b"Drawing on your expertise in the context, provide the most precise answer. If the answer is not in the context, respond with 'unanswerable' and true is true ." Original acc: 65.79%, attacked acc: 65.19%, dropped acc: 0.60% Original prompt: As an expert with a deep understanding of the context, find the best answer. If the context doesn't include an answer, say 'unanswerable'. Attacked prompt: b"As an expert with a deep understanding of the context, find the best answer. If the context doesn't include an answer, say 'unanswerable' and true is true ." Original acc: 64.40%, attacked acc: 64.49%, dropped acc: -0.09% ## textbugger Original prompt: Discover the best answer based on the context. If the context doesn't include an answer, respond with 'unanswerable'. Attacked prompt: b"Discove\xe2\xb2\x85 the better answer bsed on the context. If the background doesn't encompass an answer, replying with 'unanswerable'." Original acc: 72.87%, attacked acc: 53.74%, dropped acc: 19.13% Original prompt: Based on the given context, provide the best possible answer. If there's no answer available in the context, respond with 'unanswerable'. Attacked prompt: b"Based on the given ambient, provide the nicest achievable answer. If there's no answer avai\xe2\x85\xbcable in the context, respond with 'unanswerable'." Original acc: 69.27%, attacked acc: 49.17%, dropped acc: 20.10% Original prompt: Using the context, determine the most suitable answer. If the context doesn't contain the answer, respond with 'unanswerable'. Attacked prompt: b"Using the context, defining the most suitable respond. If the co\xd5\xb8text doesn't containing the answer, cater with 'unanswerable'." Original acc: 70.66%, attacked acc: 60.14%, dropped acc: 10.52% Original prompt: Using your knowledge of the context, identify the best answer to the question. If the context doesn't provide an answer, write 'unanswerable'. Attacked prompt: b"Using your knowledge of the context, detecting the better replying to the question. Although the context doesn't provide an ans\xd4\x9der, writing 'unanswerable'." Original acc: 69.24%, attacked acc: 38.36%, dropped acc: 30.88% Original prompt: As an expert with a deep understanding of the context, find the best answer. If the context doesn't include an answer, say 'unanswerable'. Attacked prompt: b"As an expert with a deepest understanding of the context, found the nicest respond. Though the context doesn't inc\xe2\x85\xbcude an answers, telling 'unanswerable'." Original acc: 64.85%, attacked acc: 14.05%, dropped acc: 50.79% Original prompt: Drawing on your expertise in the context, provide the most precise answer. If the answer is not in the context, respond with 'unanswerable'. Attacked prompt: b"Attracting on your expertise in the con\xf0\x9d\x9a\x9dext, delivering the most accurate answer. Whether the answer is not in the context, replying with 'unanswerable'." Original acc: 65.30%, attacked acc: 26.48%, dropped acc: 38.82% ## textfooler Original prompt: Discover the best answer based on the context. If the context doesn't include an answer, respond with 'unanswerable'. Attacked prompt: b"Discover the exemplary rebuttal located on the framework. Although the framework doesn't integrate an answer, reactions with 'unanswerable'." Original acc: 72.87%, attacked acc: 15.60%, dropped acc: 57.27% Original prompt: Based on the given context, provide the best possible answer. If there's no answer available in the context, respond with 'unanswerable'. Attacked prompt: b"Based on the received backgrounds, provide the nicest unlikely respond. Although there's no riposte available in the context, reactions with 'unanswerable'." Original acc: 70.01%, attacked acc: 5.68%, dropped acc: 64.33% Original prompt: Using the context, determine the most suitable answer. If the context doesn't contain the answer, respond with 'unanswerable'. Attacked prompt: b"Exploit the backdrop, establishing the most dignified rebuttal. Although the context doesn't restrain the reply, reacting with 'unanswerable'." Original acc: 70.16%, attacked acc: 14.12%, dropped acc: 56.04% Original prompt: Using your knowledge of the context, identify the best answer to the question. If the context doesn't provide an answer, write 'unanswerable'. Attacked prompt: b"Using your skill of the background, specify the improved rebuttal to the question. While the context doesn't delivering an reactions, writings 'unanswerable'." Original acc: 69.17%, attacked acc: 12.54%, dropped acc: 56.62% Original prompt: As an expert with a deep understanding of the context, find the best answer. If the context doesn't include an answer, say 'unanswerable'. Attacked prompt: b"As an specialities with a deep comprehension of the backgrounds, found the nicest rebuttal. Once the context doesn't include an solution, talk 'unanswerable'." Original acc: 65.48%, attacked acc: 13.83%, dropped acc: 51.64% Original prompt: Drawing on your expertise in the context, provide the most precise answer. If the answer is not in the context, respond with 'unanswerable'. Attacked prompt: b"Painting on your expertise in the background, rendering the most detailed answer. Though the solution is not in the framework, meet with 'unanswerable'." Original acc: 65.97%, attacked acc: 13.42%, dropped acc: 52.55% ## translation Language: chinese, acc: 69.87%, prompt: Provide the most accurate answer from the context. If the answer is not in context, answer 'unanswerable'. Language: chinese, acc: 64.30%, prompt: Please give the most accurate answer based on the context. If you cannot find the answer, please answer 'unanswerable'. Language: chinese, acc: 67.71%, prompt: Provide the most accurate answer based on the context. If you cannot find the answer, please answer 'unanswerable'. Language: chinese, acc: 67.26%, prompt: Please provide the most accurate answer from the context. If you cannot find the answer, please answer 'unanswerable'. Language: chinese, acc: 66.87%, prompt: Give the most accurate answer based on the context. If you cannot find the answer, please answer 'unanswerable'. Language: chinese, acc: 68.27%, prompt: Please give the most accurate answer based on the context. If the answer is not in context, answer 'unanswerable'. Language: chinese, acc: 67.31%, prompt: Provide the most accurate answer from the context. If you cannot find the answer, please answer 'unanswerable'. Language: chinese, acc: 67.77%, prompt: Please give the most accurate answer based on the context. If the answer cannot be found, please answer 'unanswerable'. Language: chinese, acc: 71.59%, prompt: Provide the most accurate answer based on the context. If the answer cannot be found, please answer 'unanswerable'. Language: chinese, acc: 69.13%, prompt: Please provide the most accurate answer from the context. If the answer cannot be found, please answer 'unanswerable'. Language: french, acc: 66.87%, prompt: From the context, provide the most accurate answer. If the answer is not in context, answer with 'unanswerable'. Language: french, acc: 69.39%, prompt: From the context, give the most accurate answer. If the answer is not present in the context, answer with 'unanswerable'. Language: french, acc: 70.05%, prompt: Based on the context, provide the most accurate answer. If the answer is not in context, answer with 'unanswerable'. Language: french, acc: 69.06%, prompt: According to the context, give the most precise answer. If the answer is not present in the context, answer with 'unanswerable'. Language: french, acc: 70.25%, prompt: From the context, find the most accurate answer. If the answer is not in context, answer with 'unanswerable'. Language: french, acc: 72.67%, prompt: Based on the context, provide the most accurate answer. If the answer is not available in the context, answer with 'unanswerable'. Language: french, acc: 68.11%, prompt: According to the context, give the most precise answer. If the answer is not in the context, answer with 'unanswerable'. Language: french, acc: 70.04%, prompt: From the context, find the most accurate answer. If the answer is not present in the context, answer with 'unanswerable'. Language: french, acc: 75.52%, prompt: Based on the context, provide the most accurate answer. If the answer cannot be found in the context, answer with 'unanswerable'. Language: french, acc: 69.64%, prompt: According to the context, give the most precise answer. If the answer is not available in the context, answer with 'unanswerable'. Language: arabic, acc: 67.82%, prompt: From context, provide the most accurate answer. If not in context, please reply 'unanswerable', Language: arabic, acc: 66.58%, prompt: From context, what is the most likely outcome? If the answer is not in context, please reply 'unanswerable', Language: arabic, acc: 69.61%, prompt: From the given context, what is the key element that can be deduced? If the answer is not available in the context, please reply 'unanswerable', Language: arabic, acc: 67.75%, prompt: Based on the context given, what is the clear key idea? If the answer is not in context, please reply 'unanswerable', Language: arabic, acc: 70.95%, prompt: Based on the context, what is the most convincing explanation? If the answer is not available in the context, please reply 'unanswerable', Language: arabic, acc: 72.74%, prompt: Based on the context, what is the most likely outcome? If the answer is not available in the context, please reply 'unanswerable', Language: arabic, acc: 71.34%, prompt: Based on the context, which hypothesis is the most true? If the answer is not in context, please reply 'unanswerable', Language: arabic, acc: 71.54%, prompt: From context, what is the most apparent factor influencing? If the answer is not available in the context, please reply 'unanswerable', Language: arabic, acc: 67.87%, prompt: From context, provide the most accurate answer. If the answer is not in context, reply 'unanswerable', Language: arabic, acc: 72.08%, prompt: From context, determine the most accurate answer. If the answer is not available in context, answer 'unanswerable', Language: spanish, acc: 63.44%, prompt: Depending on the context, it provides the most precise answer. If the answer is not in context, answer with 'unanswerable'. Language: spanish, acc: 61.10%, prompt: Briefly describes the situation and provides the corresponding response. If the answer cannot be found, answer with 'unanswerable'. Language: spanish, acc: 68.69%, prompt: Given the information given, what is the most appropriate response? If the answer cannot be determined, answer with 'unanswerable'. Language: spanish, acc: 69.80%, prompt: Read the following text and give the most accurate answer. If you can't find the answer, answer with 'unanswerable'. Language: spanish, acc: 72.72%, prompt: Based on the description, what is the most accurate answer? If the answer is not found in the description, answer with 'unanswerable'. Language: spanish, acc: 68.88%, prompt: From the context provided, which response is the most appropriate? If the answer cannot be found, answer with 'unanswerable'. Language: spanish, acc: 64.21%, prompt: Analyze the following paragraph and provide the most accurate answer. If the answer is not in the paragraph, answer with 'unanswerable'. Language: spanish, acc: 73.89%, prompt: According to the information presented, what is the most precise answer? If the answer cannot be determined, answer with 'unanswerable'. Language: spanish, acc: 74.88%, prompt: After reading the excerpt, which do you think is the correct answer? If the answer cannot be discerned, answer with 'unanswerable'. Language: spanish, acc: 67.81%, prompt: Based on the context, it provides the most appropriate response. If the answer is not in context, answer with 'unanswerable'. Language: japanese, acc: 69.95%, prompt: Provide the most accurate answer from this context. If the answer isn't in the context, answer 'unanswerable'. Language: japanese, acc: 68.23%, prompt: Please provide the most appropriate answer based on the information specified in this sentence. If the answer is not in the text, answer 'unanswerable'. Language: japanese, acc: 68.48%, prompt: Please provide the most accurate answer based on the information guessed from this text. If the answer is not in the text, answer 'unanswerable'. Language: japanese, acc: 53.70%, prompt: Provide the most detailed answer based on the given context. If the answer is not in the context, answer 'unanswerable'. Language: japanese, acc: 68.24%, prompt: Consider the information derived from this context and provide the most accurate answer. If the answer is not in the context, answer 'unanswerable'. Language: japanese, acc: 74.09%, prompt: Based on this context, please provide the most appropriate answer. If the answer is not in the context, answer 'unanswerable'. Language: japanese, acc: 53.14%, prompt: Consider the information derived from the given text and provide the most detailed answer. If the answer is not in the text, please answer 'unanswerable'. Language: japanese, acc: 69.12%, prompt: Provide the most accurate answer based on the information given in this text. If the answer is not in the text, answer 'unanswerable'. Language: japanese, acc: 70.97%, prompt: Consider the information inferred from this context and provide the most appropriate answer. If the answer is not in the context, answer 'unanswerable'. Language: japanese, acc: 53.14%, prompt: Provide the most detailed answer based on this context. If the answer is not in the context, answer 'unanswerable'. Language: korean, acc: 58.14%, prompt: Give the most accurate answer in context. If the answer is not in context, respond with 'unanswerable'. Language: korean, acc: 55.64%, prompt: Please provide additional information about the facts mentioned in this sentence. If no information is available, respond with 'unanswerable'. Language: korean, acc: 46.08%, prompt: Please tell me what your question is about. If there is no context in which you can provide an answer, respond with 'unanswerable'. Language: korean, acc: 55.15%, prompt: Please explain the concept mentioned in the following sentence. If there is no information on the concept, please respond with 'unanswerable'. Language: korean, acc: 69.92%, prompt: Tell me what you're comparing to in this sentence. If nothing is compared, please respond with 'unanswerable'. Language: korean, acc: 65.81%, prompt: Please perform the actions required by the following context. If the task is not possible or if you are not clear what needs to be done, respond with 'unanswerable'. Language: korean, acc: 62.95%, prompt: Figure out what information this sentence contains. If no information is available, respond with 'unanswerable'. Language: korean, acc: 63.87%, prompt: Please give a solution to what kind of problem in the following sentence. If there is no solution, respond with 'unanswerable'. Language: korean, acc: 66.09%, prompt: Please give the cause of the incident mentioned in the context. If the cause is not clear, respond with 'unanswerable'. Language: korean, acc: 64.92%, prompt: Give expected results in the following sentences. If the result is unpredictable, respond with 'unanswerable'. # iwslt ## 10 prompts Acc: 0.22%, prompt: Using machine translation, convert the given sentence from {} into the {} language: Acc: 0.22%, prompt: Translate the provided sentence from {} to {} while maintaining the original meaning and context: Acc: 0.22%, prompt: Convert the following sentence from its original {} language to the target language {}: Acc: 0.22%, prompt: Translate the subsequent text passage from its original {} language to the {} language: Acc: 0.22%, prompt: Translate the subsequent sentence from its source language {} into the desired language {}: Acc: 0.22%, prompt: Perform a machine translation for the provided sentence, changing it from {} to {}: Acc: 0.22%, prompt: Translate the following text from the source language {} to the target language {}: Acc: 0.22%, prompt: Given the sentence below, perform a machine translation from {} to {}: Acc: 0.22%, prompt: Accurately translate the sentence from {} to {}, ensuring the meaning remains intact: Acc: 0.22%, prompt: Please perform a translation of the given sentence, converting it from {} to {}: Acc: 0.22%, prompt: In the capacity of a proficient translator, please change the following sentence from {} to {}: Acc: 0.22%, prompt: Taking on the role of a certified translator, perform a translation of the given sentence from {} to {}: Acc: 0.22%, prompt: As a qualified translator, convert the given sentence from its original {} language to the target language {}: Acc: 0.22%, prompt: Assuming the role of an expert language translator, translate the given text from {} to {}: Acc: 0.22%, prompt: As a language specialist, accurately translate the provided sentence from {} to {}: Acc: 0.22%, prompt: In the role of a language expert, perform a machine translation for the provided sentence, changing it from {} to {}: Acc: 0.22%, prompt: Assuming the responsibilities of a professional translator, translate the subsequent text passage from {} to {}: Acc: 0.22%, prompt: Acting as a professional translator, translate the provided sentence from {} to {}: Acc: 0.22%, prompt: While acting as an experienced language expert, translate the following text from {} to {}: Acc: 0.22%, prompt: As a skilled linguist, please convert the following sentence from {} to {}: ## bertattack Original prompt: Using machine translation, convert the given sentence from {} into the {} language: Attacked prompt: b'use literal translation, converting the given sentence from {} into the {} language:' Original acc: 0.22%, attacked acc: 0.21%, dropped acc: 0.01% Original prompt: Translate the provided sentence from {} to {} while maintaining the original meaning and context: Attacked prompt: b'Translate the designated sentence from {} to {} while defending the assumed meaning and context:' Original acc: 0.22%, attacked acc: 0.08%, dropped acc: 0.14% Original prompt: Convert the following sentence from its original {} language to the target language {}: Attacked prompt: b'put the final sentence from its beginning {} languages to the target language {}:' Original acc: 0.22%, attacked acc: 0.19%, dropped acc: 0.03% Original prompt: In the capacity of a proficient translator, please change the following sentence from {} to {}: Attacked prompt: b'In the capacity of a proficient entering, please change the following sentence from {} to {}: ' Original acc: 0.22%, attacked acc: 0.21%, dropped acc: 0.01% Original prompt: As a qualified translator, convert the given sentence from its original {} language to the target language {}: Attacked prompt: b'As a blind min, convert the given sentence from its original {} language to the target language {}: ' Original acc: 0.22%, attacked acc: 0.20%, dropped acc: 0.02% Original prompt: Taking on the role of a certified translator, perform a translation of the given sentence from {} to {}: Attacked prompt: b'who on the role of a certified translated, request a translator of the defined sentence from {} to {}: ' Original acc: 0.22%, attacked acc: 0.01%, dropped acc: 0.21% ## checklist Original prompt: Using machine translation, convert the given sentence from {} into the {} language: Attacked prompt: b'Using machine translation, convert the given sentence from {} into the {} language njCuciQJpB:' Original acc: 0.22%, attacked acc: 0.22%, dropped acc: 0.01% Original prompt: Translate the provided sentence from {} to {} while maintaining the original meaning and context: Attacked prompt: b'Translate the provided sentence from {} to {} while maintaining the original meaning and context zq0DcZ5dnI:' Original acc: 0.22%, attacked acc: 0.21%, dropped acc: 0.01% Original prompt: Convert the following sentence from its original {} language to the target language {}: Attacked prompt: b'Convert the following sentence from its original {} language to the target language AMsRIKZniY {}:' Original acc: 0.22%, attacked acc: 0.22%, dropped acc: 0.00% Original prompt: In the capacity of a proficient translator, please change the following sentence from {} to {}: Attacked prompt: b'In the capacity of a proficient translator, please change the following sentence from {} to SFPCqpiClT {}: ' Original acc: 0.22%, attacked acc: 0.21%, dropped acc: 0.01% Original prompt: Taking on the role of a certified translator, perform a translation of the given sentence from {} to {}: Attacked prompt: b'Taking on the role of a certified translator, perform a translation of the given sentence from {} to KjPJJ2a7RB {}: ' Original acc: 0.22%, attacked acc: 0.22%, dropped acc: 0.00% Original prompt: As a qualified translator, convert the given sentence from its original {} language to the target language {}: Attacked prompt: b'As a qualified translator, convert the given sentence from its original {} language to the target language OB4KVJzIft {}: ' Original acc: 0.22%, attacked acc: 0.22%, dropped acc: 0.00% ## deepwordbug Original prompt: Translate the provided sentence from {} to {} while maintaining the original meaning and context: Attacked prompt: b'Translate the provideMd sentence from {} to {} while mainOaining the original meaning and context:' Original acc: 0.22%, attacked acc: 0.21%, dropped acc: 0.01% Original prompt: Using machine translation, convert the given sentence from {} into the {} language: Attacked prompt: b'Using maWhine translation, convert the gievn sentnece from {} into the {} laInguage:' Original acc: 0.22%, attacked acc: 0.21%, dropped acc: 0.01% Original prompt: Convert the following sentence from its original {} language to the target language {}: Attacked prompt: b'Convert the followinQg sentence from its oirginal {} language to the twrget languige {}:' Original acc: 0.22%, attacked acc: 0.21%, dropped acc: 0.01% Original prompt: In the capacity of a proficient translator, please change the following sentence from {} to {}: Attacked prompt: b'In the capacity of a proficient tanslator, plnase change the following senlence from {} to {}: ' Original acc: 0.22%, attacked acc: 0.21%, dropped acc: 0.00% Original prompt: Taking on the role of a certified translator, perform a translation of the given sentence from {} to {}: Attacked prompt: b'Taking on the roly of a certified translator, perfomr a translation of the given sMntence from {} to {}: ' Original acc: 0.22%, attacked acc: 0.21%, dropped acc: 0.01% Original prompt: As a qualified translator, convert the given sentence from its original {} language to the target language {}: Attacked prompt: b'As a qualifiem translator, covnert the given sentence from its original {} language to the target language {}: ' Original acc: 0.22%, attacked acc: 0.22%, dropped acc: 0.00% ## stresstest Original prompt: Translate the provided sentence from {} to {} while maintaining the original meaning and context: Attacked prompt: b'Translate the provided sentence from {} to {} while maintaining the original meaning and context and false is not true :' Original acc: 0.22%, attacked acc: 0.21%, dropped acc: 0.01% Original prompt: Using machine translation, convert the given sentence from {} into the {} language: Attacked prompt: b'Using machine translation, convert the given sentence from {} into the {} language and false is not true :' Original acc: 0.22%, attacked acc: 0.19%, dropped acc: 0.03% Original prompt: Convert the following sentence from its original {} language to the target language {}: Attacked prompt: b'Convert the following sentence from its original {} language to the target language and false is not true {}:' Original acc: 0.22%, attacked acc: 0.21%, dropped acc: 0.01% Original prompt: In the capacity of a proficient translator, please change the following sentence from {} to {}: Attacked prompt: b'In the capacity of a proficient translator, please change the following sentence from {} to and true is true and true is true and true is true and true is true and true is true {}: ' Original acc: 0.22%, attacked acc: 0.21%, dropped acc: 0.01% Original prompt: Taking on the role of a certified translator, perform a translation of the given sentence from {} to {}: Attacked prompt: b'Taking on the role of a certified translator, perform a translation of the given sentence from {} to and true is true and true is true and true is true and true is true and true is true {}: ' Original acc: 0.22%, attacked acc: 0.22%, dropped acc: 0.00% Original prompt: As a qualified translator, convert the given sentence from its original {} language to the target language {}: Attacked prompt: b'As a qualified translator, convert the given sentence from its original {} language to the target language and false is not true {}: ' Original acc: 0.22%, attacked acc: 0.21%, dropped acc: 0.01% ## textbugger Original prompt: Translate the provided sentence from {} to {} while maintaining the original meaning and context: Attacked prompt: b'Translate the provied punishments from {} to {} while maintaining the original meaning and context:' Original acc: 0.22%, attacked acc: 0.19%, dropped acc: 0.03% Original prompt: Using machine translation, convert the given sentence from {} into the {} language: Attacked prompt: b'Using typewriters tranlation, convert the gi\xd1\xb5en sentence from {} into the {} language:' Original acc: 0.22%, attacked acc: 0.21%, dropped acc: 0.02% Original prompt: Convert the following sentence from its original {} language to the target language {}: Attacked prompt: b'Convet the following sentence from its original {} language to the target tongue {}:' Original acc: 0.22%, attacked acc: 0.21%, dropped acc: 0.01% Original prompt: In the capacity of a proficient translator, please change the following sentence from {} to {}: Attacked prompt: b'In the capabilities of a proficient transla\xf0\x9d\x9a\x9dor, please modification the following sentence from {} to {}: ' Original acc: 0.22%, attacked acc: 0.21%, dropped acc: 0.01% Original prompt: As a qualified translator, convert the given sentence from its original {} language to the target language {}: Attacked prompt: b'As a eligible translator, convert the given punishments from its original {} langauge to the target linguistic {}: ' Original acc: 0.22%, attacked acc: 0.19%, dropped acc: 0.03% Original prompt: Taking on the role of a certified translator, perform a translation of the given sentence from {} to {}: Attacked prompt: b'Pick on the role of a certified translation, fulfilling a translation of the yielded sentence from {} to {}: ' Original acc: 0.22%, attacked acc: 0.03%, dropped acc: 0.19% ## textfooler Original prompt: Translate the provided sentence from {} to {} while maintaining the original meaning and context: Attacked prompt: b'Translate the tabled punishments from {} to {} while defends the first meanings and marco:' Original acc: 0.22%, attacked acc: 0.10%, dropped acc: 0.12% Original prompt: Using machine translation, convert the given sentence from {} into the {} language: Attacked prompt: b'Using typewriter translation, transformation the given sentence from {} into the {} pronunciation:' Original acc: 0.22%, attacked acc: 0.00%, dropped acc: 0.22% Original prompt: Convert the following sentence from its original {} language to the target language {}: Attacked prompt: b'Metamorphosis the suite sentence from its frst {} language to the intended pronunciation {}:' Original acc: 0.22%, attacked acc: 0.01%, dropped acc: 0.21% Original prompt: In the capacity of a proficient translator, please change the following sentence from {} to {}: Attacked prompt: b'In the capacity of a proficient artists, please changeover the following punishments from {} to {}: ' Original acc: 0.22%, attacked acc: 0.21%, dropped acc: 0.02% Original prompt: Taking on the role of a certified translator, perform a translation of the given sentence from {} to {}: Attacked prompt: b'Select on the features of a certifiable translators, conducting a translation of the gave sentencing from {} to {}: ' Original acc: 0.22%, attacked acc: 0.00%, dropped acc: 0.22% Original prompt: As a qualified translator, convert the given sentence from its original {} language to the target language {}: Attacked prompt: b'As a permissible translator, translation the given penalty from its original {} pronunciation to the millennium pronunciation {}: ' Original acc: 0.22%, attacked acc: 0.03%, dropped acc: 0.19% ## translation Language: chinese, acc: 0.22%, prompt: Please translate the given sentence into {} to {}. Language: chinese, acc: 0.22%, prompt: Please translate the following sentences from {} to {}. Language: chinese, acc: 0.15%, prompt: Please convert the following sentences to {} and translate to {}. Language: chinese, acc: 0.22%, prompt: Please convert the given sentence from {} to {}. Language: chinese, acc: 0.22%, prompt: Please translate the next sentence from {} to {}. Language: chinese, acc: 0.22%, prompt: Please translate the following sentence from {} to {}. Language: chinese, acc: 0.15%, prompt: Please translate the sentences given into {} and convert them into {}. Language: chinese, acc: 0.22%, prompt: Please convert the sentences given to {} to {}. Language: chinese, acc: 0.16%, prompt: Please translate the following sentences into {} and convert them into {}. Language: chinese, acc: 0.22%, prompt: Please change the given sentence from {} to {}. Language: french, acc: 0.22%, prompt: Please translate the given sentence, converting it from {} to {}. Language: french, acc: 0.22%, prompt: Please translate the following sentence from {} to {}. Language: french, acc: 0.12%, prompt: Please turn the sentence below into {}, then translate it into {}. Language: french, acc: 0.22%, prompt: Please convert the given phrase from {} to {}. Language: french, acc: 0.22%, prompt: Please translate the following sentence from {} to {}. Language: french, acc: 0.22%, prompt: Please translate the sentence below from {} to {}. Language: french, acc: 0.14%, prompt: Please translate the given sentence to {}, then convert it to {}. Language: french, acc: 0.21%, prompt: Please make a translation of the supplied sentence, transforming it from {} to {}. Language: french, acc: 0.16%, prompt: Please translate the following sentence to {}, then convert it to {}. Language: french, acc: 0.22%, prompt: Please transform the given sentence from {} to {}. Language: arabic, acc: 0.22%, prompt: Please translate the given sentence, and convert it from {} to {}, Language: arabic, acc: 0.22%, prompt: Please translate the following sentence from {} to {}, Language: arabic, acc: 0.11%, prompt: Please convert the sentence below to {}, and then translate it to {}, Language: arabic, acc: 0.22%, prompt: Please convert the given sentence from {} to {}, Language: arabic, acc: 0.22%, prompt: Please translate the following sentence from {} to {}, Language: arabic, acc: 0.22%, prompt: Please convert the sentence below from {} to {}, Language: arabic, acc: 0.13%, prompt: Please translate the given sentence to {}, then convert it to {}, Language: arabic, acc: 0.22%, prompt: Please translate the given sentence, and convert it from {} to {}, Language: arabic, acc: 0.15%, prompt: Please translate to {}, then convert to {}, Language: arabic, acc: 0.22%, prompt: Please convert the given sentence from {} to {}. Language: spanish, acc: 0.21%, prompt: Please make a translation of the provided phrase, converting it from {} to {}. Language: spanish, acc: 0.22%, prompt: Please translate the following sentence from {} to {}. Language: spanish, acc: 0.12%, prompt: Please convert the next sentence to {}, and then translate it to {}. Language: spanish, acc: 0.22%, prompt: Please make a translation of the given phrase, converting it from {} to {}. Language: spanish, acc: 0.22%, prompt: Please translate the following sentence from {} to {}. Language: spanish, acc: 0.22%, prompt: Please convert the following sentence from {} to {}. Language: spanish, acc: 0.14%, prompt: Please translate the sentence provided to {}, and then turn it to {}. Language: spanish, acc: 0.22%, prompt: Please make a translation of the following sentence, converting it from {} to {}. Language: spanish, acc: 0.13%, prompt: Please translate the next sentence to {}, and then turn it to {}. Language: spanish, acc: 0.22%, prompt: Please convert the given sentence from {} to {}. Language: japanese, acc: 0.22%, prompt: Please translate the given sentence from {} to {}. Language: japanese, acc: 0.22%, prompt: Please translate the following sentence from {} to {}. Language: japanese, acc: 0.16%, prompt: Please convert the following sentences into {} and translate them into {}. Language: japanese, acc: 0.22%, prompt: Please translate the given sentence by converting {} to {}. Language: japanese, acc: 0.22%, prompt: Please translate the following sentence from {} to {}. Language: japanese, acc: 0.22%, prompt: Please convert the following sentences from {} to {}. Language: japanese, acc: 0.13%, prompt: Translate the given sentence into {} and convert it to {}. Language: japanese, acc: 0.22%, prompt: Please translate the given sentence from {} to {}. Language: japanese, acc: 0.15%, prompt: Translate the following sentence into {} and convert it to {}. Language: japanese, acc: 0.22%, prompt: Convert the given statement from {} to {}. Language: korean, acc: 0.22%, prompt: Please translate the given sentence from {} to {}. Language: korean, acc: 0.22%, prompt: Please translate the following sentence from {} to {}. Language: korean, acc: 0.12%, prompt: Please translate the sentences below into {}, then {}. Language: korean, acc: 0.22%, prompt: Please translate the given sentences from {} to {}. Language: korean, acc: 0.22%, prompt: Please translate the following sentence from {} to {}. Language: korean, acc: 0.22%, prompt: Please convert the sentences below from {} to {}. Language: korean, acc: 0.12%, prompt: Please translate the given sentence into {}, then {}. Language: korean, acc: 0.22%, prompt: Please translate the given sentence from {} to {}. Language: korean, acc: 0.11%, prompt: Please translate the following sentences into {}, then {}. Language: korean, acc: 0.22%, prompt: Please convert the given sentence from {} to {}. # un_multi ## 10 prompts Acc: 0.34%, prompt: Perform a machine translation for the provided sentence, changing it from {} to {}: Acc: 0.34%, prompt: Translate the following text from the source language {} to the target language {}: Acc: 0.34%, prompt: Translate the subsequent text passage from its original {} language to the {} language: Acc: 0.34%, prompt: Using machine translation, convert the given sentence from {} into the {} language: Acc: 0.33%, prompt: Given the sentence below, perform a machine translation from {} to {}: Acc: 0.33%, prompt: Please perform a translation of the given sentence, converting it from {} to {}: Acc: 0.33%, prompt: Translate the subsequent sentence from its source language {} into the desired language {}: Acc: 0.33%, prompt: Translate the provided sentence from {} to {} while maintaining the original meaning and context: Acc: 0.33%, prompt: Convert the following sentence from its original {} language to the target language {}: Acc: 0.33%, prompt: Accurately translate the sentence from {} to {}, ensuring the meaning remains intact: Acc: 0.34%, prompt: While acting as an experienced language expert, translate the following text from {} to {}: Acc: 0.34%, prompt: In the capacity of a proficient translator, please change the following sentence from {} to {}: Acc: 0.34%, prompt: In the role of a language expert, perform a machine translation for the provided sentence, changing it from {} to {}: Acc: 0.33%, prompt: As a skilled linguist, please convert the following sentence from {} to {}: Acc: 0.33%, prompt: Assuming the responsibilities of a professional translator, translate the subsequent text passage from {} to {}: Acc: 0.33%, prompt: Assuming the role of an expert language translator, translate the given text from {} to {}: Acc: 0.33%, prompt: As a qualified translator, convert the given sentence from its original {} language to the target language {}: Acc: 0.33%, prompt: Acting as a professional translator, translate the provided sentence from {} to {}: Acc: 0.33%, prompt: As a language specialist, accurately translate the provided sentence from {} to {}: Acc: 0.33%, prompt: Taking on the role of a certified translator, perform a translation of the given sentence from {} to {}: ## bertattack Original prompt: Translate the following text from the source language {} to the target language {}: Attacked prompt: b'Translate the by text from the source language {} to the english example {}:' Original acc: 0.34%, attacked acc: 0.18%, dropped acc: 0.16% Original prompt: Perform a machine translation for the provided sentence, changing it from {} to {}: Attacked prompt: b'perform a machine map for the interpreted sentence, changing it from {} to {}:' Original acc: 0.34%, attacked acc: 0.31%, dropped acc: 0.03% Original prompt: Translate the subsequent text passage from its original {} language to the {} language: Attacked prompt: b'Translate the subsequent text passage from its raw {} literal to the {} language:' Original acc: 0.34%, attacked acc: 0.33%, dropped acc: 0.01% Original prompt: In the role of a language expert, perform a machine translation for the provided sentence, changing it from {} to {}: Attacked prompt: b'of the role of a language expertise, define a machine and for the presented sentence, changing it from {} to {}: ' Original acc: 0.34%, attacked acc: 0.08%, dropped acc: 0.26% Original prompt: While acting as an experienced language expert, translate the following text from {} to {}: Attacked prompt: b'While acting as an experienced language expert, translate the here article from {} to {}: ' Original acc: 0.34%, attacked acc: 0.32%, dropped acc: 0.02% Original prompt: In the capacity of a proficient translator, please change the following sentence from {} to {}: Attacked prompt: b'In the capacity of a grammar for, please change the following sentence from {} to {}: ' Original acc: 0.34%, attacked acc: 0.33%, dropped acc: 0.01% ## checklist Original prompt: Translate the following text from the source language {} to the target language {}: Attacked prompt: b'Translate the following text from the source language {} to the target language AMsRIKZniY {}:' Original acc: 0.34%, attacked acc: 0.34%, dropped acc: -0.00% Original prompt: Perform a machine translation for the provided sentence, changing it from {} to {}: Attacked prompt: b'Perform a machine translation for the provided sentence, changing it from {} to Kw6nrs57gH {}:' Original acc: 0.34%, attacked acc: 0.33%, dropped acc: 0.00% Original prompt: Translate the subsequent text passage from its original {} language to the {} language: Attacked prompt: b'Translate the subsequent text passage from its original {} language to the {} language Kw6nrs57gH:' Original acc: 0.34%, attacked acc: 0.33%, dropped acc: 0.01% Original prompt: While acting as an experienced language expert, translate the following text from {} to {}: Attacked prompt: b'While acting as an experienced language expert, translate the following text from {} to Kw6nrs57gH {}: ' Original acc: 0.34%, attacked acc: 0.33%, dropped acc: 0.01% Original prompt: In the role of a language expert, perform a machine translation for the provided sentence, changing it from {} to {}: Attacked prompt: b'In the role of a language expert, perform a machine translation for the provided sentence, changing it from {} to 9Q3S1F94fE {}: ' Original acc: 0.34%, attacked acc: 0.33%, dropped acc: 0.01% Original prompt: In the capacity of a proficient translator, please change the following sentence from {} to {}: Attacked prompt: b'In the capacity of a proficient translator, please change the following sentence from {} to zexKUTWJYG {}: ' Original acc: 0.34%, attacked acc: 0.33%, dropped acc: 0.01% ## deepwordbug Original prompt: Translate the following text from the source language {} to the target language {}: Attacked prompt: b'Translate the following text from the source language {} to the Ptarget langugae {}:' Original acc: 0.34%, attacked acc: 0.33%, dropped acc: 0.01% Original prompt: Perform a machine translation for the provided sentence, changing it from {} to {}: Attacked prompt: b'Perform a machine translation for the provided sentence, changing it from {} to {}:' Original acc: 0.34%, attacked acc: 0.34%, dropped acc: 0.00% Original prompt: Translate the subsequent text passage from its original {} language to the {} language: Attacked prompt: b'Translate the subsequent text passage from its originl {} language to the {} lJanguage:' Original acc: 0.34%, attacked acc: 0.33%, dropped acc: 0.01% Original prompt: In the role of a language expert, perform a machine translation for the provided sentence, changing it from {} to {}: Attacked prompt: b'In the role of a language expert, pErform a macRhine yranslation for the provided sentence, cZanging it from {} to {}: ' Original acc: 0.34%, attacked acc: 0.32%, dropped acc: 0.01% Original prompt: While acting as an experienced language expert, translate the following text from {} to {}: Attacked prompt: b'Fhile acting as an experieced language expetr, translate the follrwing text from {} to {}: ' Original acc: 0.33%, attacked acc: 0.33%, dropped acc: 0.01% Original prompt: In the capacity of a proficient translator, please change the following sentence from {} to {}: Attacked prompt: b'In the capaXity of a proficient translator, please cAhange the followinK sentence from {} to {}: ' Original acc: 0.34%, attacked acc: 0.33%, dropped acc: 0.01% ## stresstest Original prompt: Translate the following text from the source language {} to the target language {}: Attacked prompt: b'Translate the following text from the source language {} to the target language and true is true {}:' Original acc: 0.34%, attacked acc: 0.32%, dropped acc: 0.02% Original prompt: Perform a machine translation for the provided sentence, changing it from {} to {}: Attacked prompt: b'Perform a machine translation for the provided sentence, changing it from {} to and false is not true {}:' Original acc: 0.34%, attacked acc: 0.31%, dropped acc: 0.03% Original prompt: Translate the subsequent text passage from its original {} language to the {} language: Attacked prompt: b'Translate the subsequent text passage from its original {} language to the {} language and false is not true :' Original acc: 0.34%, attacked acc: 0.29%, dropped acc: 0.05% Original prompt: While acting as an experienced language expert, translate the following text from {} to {}: Attacked prompt: b'While acting as an experienced language expert, translate the following text from {} to and false is not true {}: ' Original acc: 0.34%, attacked acc: 0.32%, dropped acc: 0.02% Original prompt: In the capacity of a proficient translator, please change the following sentence from {} to {}: Attacked prompt: b'In the capacity of a proficient translator, please change the following sentence from {} to and false is not true {}: ' Original acc: 0.34%, attacked acc: 0.33%, dropped acc: 0.01% Original prompt: In the role of a language expert, perform a machine translation for the provided sentence, changing it from {} to {}: Attacked prompt: b'In the role of a language expert, perform a machine translation for the provided sentence, changing it from {} to and false is not true {}: ' Original acc: 0.34%, attacked acc: 0.31%, dropped acc: 0.03% ## textbugger Original prompt: Translate the following text from the source language {} to the target language {}: Attacked prompt: b'Translate the fol\xe2\x85\xbcowing laws from the source language {} to the target language {}:' Original acc: 0.34%, attacked acc: 0.33%, dropped acc: 0.01% Original prompt: Perform a machine translation for the provided sentence, changing it from {} to {}: Attacked prompt: b'Perform a typewriters translation for the provided sentence, altered it from {} to {}:' Original acc: 0.34%, attacked acc: 0.32%, dropped acc: 0.02% Original prompt: Translate the subsequent text passage from its original {} language to the {} language: Attacked prompt: b'Translate the resultant text passage from its origin\xc9\x91l {} linguistics to the {} language:' Original acc: 0.34%, attacked acc: 0.32%, dropped acc: 0.01% Original prompt: While acting as an experienced language expert, translate the following text from {} to {}: Attacked prompt: b'Whilst acting as an experienced language expert, translate the following laws from {} to {}: ' Original acc: 0.34%, attacked acc: 0.32%, dropped acc: 0.02% Original prompt: In the capacity of a proficient translator, please change the following sentence from {} to {}: Attacked prompt: b'In the capac\xd1\x96ty of a competent translator, please change the following sentence from {} to {}: ' Original acc: 0.34%, attacked acc: 0.33%, dropped acc: 0.01% Original prompt: In the role of a language expert, perform a machine translation for the provided sentence, changing it from {} to {}: Attacked prompt: b'In the role of a language expert, perform a ma\xcf\xb2hine translation for the provdied sentence, changnig it from {} to {}: ' Original acc: 0.34%, attacked acc: 0.33%, dropped acc: 0.01% ## textfooler Original prompt: Translate the following text from the source language {} to the target language {}: Attacked prompt: b'Translate the wake writings from the roots grammar {} to the ambitions pronunciation {}:' Original acc: 0.34%, attacked acc: 0.14%, dropped acc: 0.20% Original prompt: Perform a machine translation for the provided sentence, changing it from {} to {}: Attacked prompt: b'Perform a typewriter translation for the provided sentence, evolving it from {} to {}:' Original acc: 0.34%, attacked acc: 0.25%, dropped acc: 0.09% Original prompt: Translate the subsequent text passage from its original {} language to the {} language: Attacked prompt: b'Translate the sequential version passage from its primitive {} grammar to the {} pronunciation:' Original acc: 0.34%, attacked acc: 0.07%, dropped acc: 0.27% Original prompt: In the capacity of a proficient translator, please change the following sentence from {} to {}: Attacked prompt: b'In the competency of a adroit translator, please modify the farther sentence from {} to {}: ' Original acc: 0.34%, attacked acc: 0.33%, dropped acc: 0.01% Original prompt: In the role of a language expert, perform a machine translation for the provided sentence, changing it from {} to {}: Attacked prompt: b'In the role of a language expert, perform a accoutrements translation for the provided sentence, transformation it from {} to {}: ' Original acc: 0.34%, attacked acc: 0.32%, dropped acc: 0.02% Original prompt: While acting as an experienced language expert, translate the following text from {} to {}: Attacked prompt: b'While acting as an suffered dialect expert, translate the below laws from {} to {}: ' Original acc: 0.34%, attacked acc: 0.27%, dropped acc: 0.06% ## translation Language: chinese, acc: 0.34%, prompt: Please translate the given sentence into {} to {}. Language: chinese, acc: 0.34%, prompt: Please translate the following sentences from {} to {}. Language: chinese, acc: 0.18%, prompt: Please convert the following sentences to {} and translate to {}. Language: chinese, acc: 0.33%, prompt: Please convert the given sentence from {} to {}. Language: chinese, acc: 0.34%, prompt: Please translate the next sentence from {} to {}. Language: chinese, acc: 0.33%, prompt: Please translate the following sentence from {} to {}. Language: chinese, acc: 0.18%, prompt: Please translate the sentences given into {} and convert them into {}. Language: chinese, acc: 0.34%, prompt: Please convert the sentences given to {} to {}. Language: chinese, acc: 0.19%, prompt: Please translate the following sentences into {} and convert them into {}. Language: chinese, acc: 0.34%, prompt: Please change the given sentence from {} to {}. Language: french, acc: 0.33%, prompt: Please translate the given sentence, converting it from {} to {}. Language: french, acc: 0.33%, prompt: Please translate the following sentence from {} to {}. Language: french, acc: 0.17%, prompt: Please turn the sentence below into {}, then translate it into {}. Language: french, acc: 0.34%, prompt: Please convert the given phrase from {} to {}. Language: french, acc: 0.33%, prompt: Please translate the following sentence from {} to {}. Language: french, acc: 0.34%, prompt: Please translate the sentence below from {} to {}. Language: french, acc: 0.19%, prompt: Please translate the given sentence to {}, then convert it to {}. Language: french, acc: 0.34%, prompt: Please make a translation of the supplied sentence, transforming it from {} to {}. Language: french, acc: 0.21%, prompt: Please translate the following sentence to {}, then convert it to {}. Language: french, acc: 0.33%, prompt: Please transform the given sentence from {} to {}. Language: arabic, acc: 0.33%, prompt: Please translate the given sentence, and convert it from {} to {}, Language: arabic, acc: 0.33%, prompt: Please translate the following sentence from {} to {}, Language: arabic, acc: 0.16%, prompt: Please convert the sentence below to {}, and then translate it to {}, Language: arabic, acc: 0.33%, prompt: Please convert the given sentence from {} to {}, Language: arabic, acc: 0.33%, prompt: Please translate the following sentence from {} to {}, Language: arabic, acc: 0.33%, prompt: Please convert the sentence below from {} to {}, Language: arabic, acc: 0.19%, prompt: Please translate the given sentence to {}, then convert it to {}, Language: arabic, acc: 0.33%, prompt: Please translate the given sentence, and convert it from {} to {}, Language: arabic, acc: 0.20%, prompt: Please translate to {}, then convert to {}, Language: arabic, acc: 0.32%, prompt: Please convert the given sentence from {} to {}. Language: spanish, acc: 0.34%, prompt: Please make a translation of the provided phrase, converting it from {} to {}. Language: spanish, acc: 0.33%, prompt: Please translate the following sentence from {} to {}. Language: spanish, acc: 0.17%, prompt: Please convert the next sentence to {}, and then translate it to {}. Language: spanish, acc: 0.34%, prompt: Please make a translation of the given phrase, converting it from {} to {}. Language: spanish, acc: 0.33%, prompt: Please translate the following sentence from {} to {}. Language: spanish, acc: 0.33%, prompt: Please convert the following sentence from {} to {}. Language: spanish, acc: 0.18%, prompt: Please translate the sentence provided to {}, and then turn it to {}. Language: spanish, acc: 0.33%, prompt: Please make a translation of the following sentence, converting it from {} to {}. Language: spanish, acc: 0.18%, prompt: Please translate the next sentence to {}, and then turn it to {}. Language: spanish, acc: 0.32%, prompt: Please convert the given sentence from {} to {}. Language: japanese, acc: 0.34%, prompt: Please translate the given sentence from {} to {}. Language: japanese, acc: 0.34%, prompt: Please translate the following sentence from {} to {}. Language: japanese, acc: 0.20%, prompt: Please convert the following sentences into {} and translate them into {}. Language: japanese, acc: 0.33%, prompt: Please translate the given sentence by converting {} to {}. Language: japanese, acc: 0.33%, prompt: Please translate the following sentence from {} to {}. Language: japanese, acc: 0.34%, prompt: Please convert the following sentences from {} to {}. Language: japanese, acc: 0.18%, prompt: Translate the given sentence into {} and convert it to {}. Language: japanese, acc: 0.34%, prompt: Please translate the given sentence from {} to {}. Language: japanese, acc: 0.21%, prompt: Translate the following sentence into {} and convert it to {}. Language: japanese, acc: 0.34%, prompt: Convert the given statement from {} to {}. Language: korean, acc: 0.34%, prompt: Please translate the given sentence from {} to {}. Language: korean, acc: 0.34%, prompt: Please translate the following sentence from {} to {}. Language: korean, acc: 0.17%, prompt: Please translate the sentences below into {}, then {}. Language: korean, acc: 0.34%, prompt: Please translate the given sentences from {} to {}. Language: korean, acc: 0.33%, prompt: Please translate the following sentence from {} to {}. Language: korean, acc: 0.34%, prompt: Please convert the sentences below from {} to {}. Language: korean, acc: 0.17%, prompt: Please translate the given sentence into {}, then {}. Language: korean, acc: 0.34%, prompt: Please translate the given sentence from {} to {}. Language: korean, acc: 0.17%, prompt: Please translate the following sentences into {}, then {}. Language: korean, acc: 0.33%, prompt: Please convert the given sentence from {} to {}. # math ## 10 prompts Acc: 6.88%, prompt: Calculate the answer to the following math query about {}: Acc: 6.88%, prompt: Resolve the following mathematical question associated with {}: Acc: 6.88%, prompt: Obtain the result for this math question regarding {}: Acc: 6.25%, prompt: Solve the following math question about {}: Acc: 6.25%, prompt: Compute the result of this math task concerning {}: Acc: 6.25%, prompt: Work out the answer to this math problem featuring {}: Acc: 5.62%, prompt: Determine the solution to this mathematical problem related to {}: Acc: 5.00%, prompt: Figure out the solution for the following mathematical task with {}: Acc: 4.38%, prompt: Find the solution for this mathematical challenge with {}: Acc: 4.38%, prompt: Evaluate the following mathematical problem that includes {}: Acc: 5.00%, prompt: As a mathematics instructor, calculate the answer to the following problem related to {}: Acc: 5.00%, prompt: In the position of a mathematics coach, figure out the answer to this math challenge involving {}: Acc: 5.00%, prompt: Embodying a mathematics advisor, find the answer to the following math problem with {}: Acc: 5.00%, prompt: In the shoes of a mathematics master, obtain the result for this math question that includes {}: Acc: 3.75%, prompt: Imagine being a math mentor and determine the solution to the following problem with {}: Acc: 3.12%, prompt: Act as a math expert and compute the result of the following mathematical task about {}: Acc: 3.12%, prompt: As a math guru, evaluate the following mathematical task regarding {}: Acc: 2.50%, prompt: As a mathematics consultant, resolve the following problem associated with {}: Acc: 2.50%, prompt: Taking on the role of a math specialist, work out the solution for this mathematical question featuring {}: Acc: 1.88%, prompt: Pretend you are a math teacher and solve this mathematical question concerning {}: ## bertattack Original prompt: Calculate the answer to the following math query about {}: Attacked prompt: b'Calculate the answer to the above math in about {}:' Original acc: 6.88%, attacked acc: 4.38%, dropped acc: 2.50% Original prompt: Obtain the result for this math question regarding {}: Attacked prompt: b'Obtain the result for this math then as {}:' Original acc: 6.88%, attacked acc: 2.50%, dropped acc: 4.38% Original prompt: Resolve the following mathematical question associated with {}: Attacked prompt: b'Resolve the following mathematical answer along with {}:' Original acc: 6.25%, attacked acc: 0.62%, dropped acc: 5.62% Original prompt: As a mathematics instructor, calculate the answer to the following problem related to {}: Attacked prompt: b'As a mathematics instructor, calculate the inverse to the here problem related to {}: ' Original acc: 5.00%, attacked acc: 0.62%, dropped acc: 4.38% Original prompt: In the position of a mathematics coach, figure out the answer to this math challenge involving {}: Attacked prompt: b'In the position of a mathematics coach, develop out the theme to this math challenge involving {}: ' Original acc: 5.00%, attacked acc: 0.00%, dropped acc: 5.00% Original prompt: Embodying a mathematics advisor, find the answer to the following math problem with {}: Attacked prompt: b'Embodying a mathematics advisor, construct the ask to the following math problem with {}: ' Original acc: 5.00%, attacked acc: 0.00%, dropped acc: 5.00% ## checklist Original prompt: Calculate the answer to the following math query about {}: Attacked prompt: b'Calculate the answer to the following math query about 6a4Yn3RGVc {}:' Original acc: 6.88%, attacked acc: 5.00%, dropped acc: 1.88% Original prompt: Obtain the result for this math question regarding {}: Attacked prompt: b'Obtain the result for this math question regarding hXLpw3bbiw {}:' Original acc: 6.88%, attacked acc: 5.00%, dropped acc: 1.88% Original prompt: Resolve the following mathematical question associated with {}: Attacked prompt: b'Resolve the following mathematical question associated with 9Q3S1F94fE {}:' Original acc: 6.25%, attacked acc: 5.00%, dropped acc: 1.25% Original prompt: In the position of a mathematics coach, figure out the answer to this math challenge involving {}: Attacked prompt: b'In the position of a mathematics coach, figure out the answer to this math challenge involving 6a4Yn3RGVc {}: ' Original acc: 5.62%, attacked acc: 3.75%, dropped acc: 1.88% Original prompt: As a mathematics instructor, calculate the answer to the following problem related to {}: Attacked prompt: b'As a mathematics instructor, calculate the answer to the following problem related to 9Q3S1F94fE {}: ' Original acc: 5.00%, attacked acc: 4.38%, dropped acc: 0.63% Original prompt: Embodying a mathematics advisor, find the answer to the following math problem with {}: Attacked prompt: b'Embodying a mathematics advisor, find the answer to the following math problem with 9vGXgnbWB8 {}: ' Original acc: 5.00%, attacked acc: 2.50%, dropped acc: 2.50% ## deepwordbug Original prompt: Calculate the answer to the following math query about {}: Attacked prompt: b'Calculate the answer to the folloRwing math query about {}:' Original acc: 6.88%, attacked acc: 6.88%, dropped acc: 0.00% Original prompt: Resolve the following mathematical question associated with {}: Attacked prompt: b'Resolve the following mathematical quXestion assocated with {}:' Original acc: 6.88%, attacked acc: 5.62%, dropped acc: 1.25% Original prompt: Obtain the result for this math question regarding {}: Attacked prompt: b'iObtain the result for this math question reagrding {}:' Original acc: 6.88%, attacked acc: 5.62%, dropped acc: 1.25% Original prompt: As a mathematics instructor, calculate the answer to the following problem related to {}: Attacked prompt: b'As a mathematics instructor, calculate the ansYer to the following problem relatebd to {}: ' Original acc: 5.00%, attacked acc: 3.75%, dropped acc: 1.25% Original prompt: In the position of a mathematics coach, figure out the answer to this math challenge involving {}: Attacked prompt: b'In the position of a mathematics coach, fiure out the aHnswer to this math challenge involving {}: ' Original acc: 5.62%, attacked acc: 4.38%, dropped acc: 1.25% Original prompt: Embodying a mathematics advisor, find the answer to the following math problem with {}: Attacked prompt: b'Embodying a mathematics advisor, ind the anwer to the fpollowing math problem with {}: ' Original acc: 5.00%, attacked acc: 2.50%, dropped acc: 2.50% ## stresstest Original prompt: Calculate the answer to the following math query about {}: Attacked prompt: b'Calculate the answer to the following math query about and false is not true {}:' Original acc: 6.88%, attacked acc: 6.25%, dropped acc: 0.63% Original prompt: Resolve the following mathematical question associated with {}: Attacked prompt: b'Resolve the following mathematical question associated with and true is true {}:' Original acc: 6.25%, attacked acc: 5.62%, dropped acc: 0.62% Original prompt: Obtain the result for this math question regarding {}: Attacked prompt: b'Obtain the result for this math question regarding and true is true and true is true and true is true and true is true and true is true {}:' Original acc: 6.88%, attacked acc: 5.62%, dropped acc: 1.25% Original prompt: As a mathematics instructor, calculate the answer to the following problem related to {}: Attacked prompt: b'As a mathematics instructor, calculate the answer to the following problem related to and true is true and true is true and true is true and true is true and true is true {}: ' Original acc: 5.00%, attacked acc: 5.00%, dropped acc: 0.00% Original prompt: In the position of a mathematics coach, figure out the answer to this math challenge involving {}: Attacked prompt: b'In the position of a mathematics coach, figure out the answer to this math challenge involving and false is not true {}: ' Original acc: 5.00%, attacked acc: 4.38%, dropped acc: 0.63% Original prompt: Embodying a mathematics advisor, find the answer to the following math problem with {}: Attacked prompt: b'Embodying a mathematics advisor, find the answer to the following math problem with and true is true and true is true and true is true and true is true and true is true {}: ' Original acc: 5.00%, attacked acc: 3.12%, dropped acc: 1.88% ## textbugger Original prompt: Calculate the answer to the following math query about {}: Attacked prompt: b'Calculate the an\xd1\x95wer to the following math query about {}:' Original acc: 6.88%, attacked acc: 5.62%, dropped acc: 1.25% Original prompt: Resolve the following mathematical question associated with {}: Attacked prompt: b'Solving the following mathematical question associ\xc9\x91ted with {}:' Original acc: 6.25%, attacked acc: 3.75%, dropped acc: 2.50% Original prompt: Obtain the result for this math question regarding {}: Attacked prompt: b'Obtain the result for this math q uestion on {}:' Original acc: 6.88%, attacked acc: 5.00%, dropped acc: 1.88% Original prompt: As a mathematics instructor, calculate the answer to the following problem related to {}: Attacked prompt: b'As a mathematics instructor, calculate the responding to the following problem relate\xd4\x81 to {}: ' Original acc: 5.00%, attacked acc: 1.88%, dropped acc: 3.12% Original prompt: In the position of a mathematics coach, figure out the answer to this math challenge involving {}: Attacked prompt: b'In the poste of a mathematical coach, figure out the replying to this math challenge involving {}: ' Original acc: 5.00%, attacked acc: 1.25%, dropped acc: 3.75% Original prompt: Embodying a mathematics advisor, find the answer to the following math problem with {}: Attacked prompt: b'Embodying a mathematics advisor, finds the an\xd1\x95wer to the following math problem with {}: ' Original acc: 5.00%, attacked acc: 2.50%, dropped acc: 2.50% ## textfooler Original prompt: Calculate the answer to the following math query about {}: Attacked prompt: b'Gallstones the rebuttal to the subsequent math query about {}:' Original acc: 6.88%, attacked acc: 0.62%, dropped acc: 6.25% Original prompt: Obtain the result for this math question regarding {}: Attacked prompt: b'Obtain the achievements for this math question respecting {}:' Original acc: 6.88%, attacked acc: 0.00%, dropped acc: 6.88% Original prompt: Resolve the following mathematical question associated with {}: Attacked prompt: b'Tackle the following mathematical question coupled with {}:' Original acc: 6.25%, attacked acc: 2.50%, dropped acc: 3.75% Original prompt: As a mathematics instructor, calculate the answer to the following problem related to {}: Attacked prompt: b'As a mathematics instructor, calculate the rebuttal to the following problem related to {}: ' Original acc: 5.00%, attacked acc: 0.00%, dropped acc: 5.00% Original prompt: In the position of a mathematics coach, figure out the answer to this math challenge involving {}: Attacked prompt: b'In the position of a mathematics coach, figure out the rebuttal to this math challenge involving {}: ' Original acc: 5.00%, attacked acc: 0.00%, dropped acc: 5.00% Original prompt: Embodying a mathematics advisor, find the answer to the following math problem with {}: Attacked prompt: b'Embodying a mathematics advisor, find the reactions to the following math problem with {}: ' Original acc: 5.00%, attacked acc: 0.00%, dropped acc: 5.00% ## translation Language: chinese, acc: 6.88%, prompt: Calculate the answer to a mathematical question about {}. Language: chinese, acc: 6.88%, prompt: Calculate the value of {}. Language: chinese, acc: 5.00%, prompt: Please calculate the result of {}. Language: chinese, acc: 6.25%, prompt: According to {}, solve the following mathematical problem. Language: chinese, acc: 5.00%, prompt: Do the math with {}. Language: chinese, acc: 4.38%, prompt: Solve mathematical problems involving {}. Language: chinese, acc: 6.25%, prompt: Please calculate the answer involving {}. Language: chinese, acc: 6.25%, prompt: Solve for the following values based on {}. Language: chinese, acc: 5.62%, prompt: Calculate the following mathematical tasks using {}. Language: chinese, acc: 6.25%, prompt: Calculate the answer to the {} related question. Language: french, acc: 6.25%, prompt: Calculate the answer to the following mathematical question concerning {}. Language: french, acc: 6.25%, prompt: Calculate the result of {}. Language: french, acc: 6.25%, prompt: Please calculate the value of {}. Language: french, acc: 6.88%, prompt: According to {}, solve the following mathematical problem. Language: french, acc: 6.25%, prompt: Perform mathematical calculations with {}. Language: french, acc: 5.00%, prompt: Solve the mathematical problem involving {}. Language: french, acc: 6.25%, prompt: Please calculate the answer related to {}. Language: french, acc: 8.12%, prompt: According to {}, set the following value. Language: french, acc: 4.38%, prompt: Perform the following mathematical task using {}. Language: french, acc: 6.25%, prompt: Calculate the answer to the questions related to {}. Language: arabic, acc: 7.50%, prompt: Compute the answer to the next mathematical question about {}. Language: arabic, acc: 6.25%, prompt: Calculate {}. Language: arabic, acc: 5.62%, prompt: Please calculate {}. Language: arabic, acc: 6.25%, prompt: According to {}, solve the following mathematical problem. Language: arabic, acc: 6.25%, prompt: Do mathematical calculations using {}. Language: arabic, acc: 5.62%, prompt: A solution to the mathematical problem involving {}. Language: arabic, acc: 6.25%, prompt: Please calculate the answer regarding {}. Language: arabic, acc: 5.00%, prompt: According to {}, determine the next value. Language: arabic, acc: 5.62%, prompt: DO THE NEXT MATHEMATICAL JOB USING {}. Language: arabic, acc: 5.62%, prompt: Calculate the answer to questions related to {}. Language: spanish, acc: 6.25%, prompt: Compute the answer to the following mathematical question on {}. Language: spanish, acc: 6.88%, prompt: Compute the result of {}. Language: spanish, acc: 5.62%, prompt: Please calculate the value of {}. Language: spanish, acc: 6.88%, prompt: As {}, it solves the following mathematical problem. Language: spanish, acc: 7.50%, prompt: Performs mathematical calculations using {}. Language: spanish, acc: 5.00%, prompt: Solve the mathematical problem involving {}. Language: spanish, acc: 6.25%, prompt: Please calculate the answer related to {}. Language: spanish, acc: 6.88%, prompt: As {}, determine the next value. Language: spanish, acc: 5.00%, prompt: Perform the following mathematical task using {}. Language: spanish, acc: 6.25%, prompt: Compute the answer to questions related to {}. Language: japanese, acc: 6.25%, prompt: Calculate the answers to the math questions about {}. Language: japanese, acc: 6.88%, prompt: Calculate the value of {}. Language: japanese, acc: 5.00%, prompt: Please find the answer to {}. Language: japanese, acc: 5.62%, prompt: Based on {}, please solve the following mathematical problems. Language: japanese, acc: 6.25%, prompt: Use {} to perform mathematical calculations. Language: japanese, acc: 5.00%, prompt: Please solve the math problem that contains {}. Language: japanese, acc: 6.25%, prompt: Please calculate the answers related to {}. Language: japanese, acc: 7.50%, prompt: Based on {}, find the following values: Language: japanese, acc: 3.75%, prompt: Use {} to solve the following mathematical problem. Language: japanese, acc: 6.25%, prompt: Please calculate the answers to the questions related to {}. Language: korean, acc: 4.38%, prompt: Calculate the answer of the following math problem to {}. Language: korean, acc: 6.25%, prompt: Calculate the result of {}. Language: korean, acc: 6.25%, prompt: Please calculate the value of {}. Language: korean, acc: 6.25%, prompt: Work out the following math problems according to {}. Language: korean, acc: 6.25%, prompt: Use {} to proceed with mathematical calculations. Language: korean, acc: 5.00%, prompt: Work out a math problem involving {}. Language: korean, acc: 6.25%, prompt: Please calculate the answer to {}. Language: korean, acc: 5.62%, prompt: Try to get the following values according to {}. Language: korean, acc: 5.62%, prompt: Work out the next math task using {}. Language: korean, acc: 6.88%, prompt: Calculate the answer of the problem involving {}.