# t5_fewshot # cola ## 10 prompts Acc: 78.10%, prompt: Assess the following sentence and determine if it is grammatically correct. Respond with 'Acceptable' or 'Unacceptable'. Acc: 77.90%, prompt: Examine the given sentence and decide if it is grammatically sound. Answer with either 'Acceptable' or 'Unacceptable'. Acc: 77.80%, prompt: Is the provided sentence grammatically correct? Respond with 'Acceptable' or 'Unacceptable'. Acc: 77.20%, prompt: Analyze the provided sentence and classify its grammatical correctness as 'Acceptable' or 'Unacceptable'. Acc: 77.20%, prompt: Determine if the grammar of the given sentence is 'Acceptable' or 'Unacceptable'. Acc: 76.90%, prompt: Please evaluate the grammatical structure of the provided sentence and answer with 'Acceptable' or 'Unacceptable'. Acc: 76.80%, prompt: Review the sentence below and identify whether its grammar is 'Acceptable' or 'Unacceptable'. Acc: 76.70%, prompt: Assess the grammatical structure of the given sentence and classify it as 'Acceptable' or 'Unacceptable'. Acc: 76.50%, prompt: Check the grammar of the following sentence and indicate if it is 'Acceptable' or 'Unacceptable'. Acc: 76.50%, prompt: Examine the sentence and decide if its grammar is 'Acceptable' or 'Unacceptable'. Acc: 76.40%, 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: 75.80%, 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: 75.70%, prompt: While performing grammar analysis, classify the grammar of the following sentence as 'acceptable' for correct grammar or 'unacceptable' for incorrect grammar. Acc: 75.50%, 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: 75.40%, prompt: Functioning as a grammar evaluation tool, analyze the given sentence and decide if it is grammatically correct, responding with 'acceptable' or 'unacceptable'. Acc: 75.30%, prompt: As a tool for determining grammatical correctness, review the sentence and categorize its grammar as either 'acceptable' or 'unacceptable'. Acc: 75.20%, prompt: As a grammar identification system, examine the provided sentence and respond with 'acceptable' for grammatically correct sentences or 'unacceptable' for incorrect ones. Acc: 75.20%, 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: 75.10%, 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: 74.50%, 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. ## bertattack Original prompt: Assess the following sentence and determine if it is grammatically correct. Respond with 'Acceptable' or 'Unacceptable'. Attacked prompt: b"Assess the following conversation and pick if it is grammatical: correct. interact with 'Acceptable' or 'Unacceptable'. " Original acc: 78.10%, attacked acc: 69.90%, dropped acc: 8.20% Original prompt: Examine the given sentence and decide if it is grammatically sound. Answer with either 'Acceptable' or 'Unacceptable'. Attacked prompt: b"Examine the formed sentence and be if it is grammaticalnot sound. answer with el 'Acceptable' or 'Unacceptable'. " Original acc: 77.90%, attacked acc: 71.00%, dropped acc: 6.90% Original prompt: Is the provided sentence grammatically correct? Respond with 'Acceptable' or 'Unacceptable'. Attacked prompt: b"does the provided plural grammatical: correct? start with 'Acceptable' or 'Unacceptable'. " Original acc: 77.80%, attacked acc: 73.40%, dropped acc: 4.40% 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 mood highlighting instrument, determine if the provided category is grammatical- correct, out with 'acceptable' for correct grammar or 'unacceptable' for incorrect math. " Original acc: 76.40%, attacked acc: 70.80%, dropped acc: 5.60% Original 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. Attacked prompt: b"As an item for grammar acquisition, determine the conversation and determine if its grammar is okay, responding with 'acceptable' for normal grammar or 'unacceptable' for using spelling. " Original acc: 75.80%, attacked acc: 69.20%, dropped acc: 6.60% 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 applying spell analysis, write the grammar of the previous phrase as 'acceptable' for so grammar or 'unacceptable' for unclear tense. " Original acc: 75.70%, attacked acc: 69.50%, dropped acc: 6.20% ## checklist Original prompt: Assess the following sentence and determine if it is grammatically correct. Respond with 'Acceptable' or 'Unacceptable'. Attacked prompt: b"Assess the following sentence and determine if it is grammatically correct. Respond with 'Acceptable' or 'Unacceptable' zexKUTWJYG. " Original acc: 78.10%, attacked acc: 77.00%, dropped acc: 1.10% Original prompt: Examine the given sentence and decide if it is grammatically sound. Answer with either 'Acceptable' or 'Unacceptable'. Attacked prompt: b"Examine the given sentence and decide if it is grammatically sound. Answer with either 'Acceptable' or 'Unacceptable' BWmsLx8xOA. " Original acc: 77.90%, attacked acc: 77.70%, dropped acc: 0.20% Original prompt: Is the provided sentence grammatically correct? Respond with 'Acceptable' or 'Unacceptable'. Attacked prompt: b"Is the provided sentence grammatically correct? Respond with 'Acceptable' or 'Unacceptable' ItGDrrA1Sc. " Original acc: 77.80%, attacked acc: 77.70%, dropped acc: 0.10% 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 zexKUTWJYG. " Original acc: 76.40%, attacked acc: 75.90%, dropped acc: 0.50% Original 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. Attacked prompt: b"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 zexKUTWJYG. " Original acc: 75.80%, attacked acc: 74.90%, dropped acc: 0.90% 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: 75.70%, attacked acc: 74.60%, dropped acc: 1.10% ## deepwordbug Original prompt: Assess the following sentence and determine if it is grammatically correct. Respond with 'Acceptable' or 'Unacceptable'. Attacked prompt: b"Assess the following sentence and dteermine if it is rgammatically Icorrect. Respond with 'Acceptable' or 'Unacceptable'. " Original acc: 78.10%, attacked acc: 72.40%, dropped acc: 5.70% Original prompt: Examine the given sentence and decide if it is grammatically sound. Answer with either 'Acceptable' or 'Unacceptable'. Attacked prompt: b"Examien the given senetnce and decide if it is rgammatically suond. Answer with eiither 'Acceptable' or 'Unacceptable'. " Original acc: 77.90%, attacked acc: 70.20%, dropped acc: 7.70% Original prompt: Is the provided sentence grammatically correct? Respond with 'Acceptable' or 'Unacceptable'. Attacked prompt: b"Is the provided semtence rammatically corrct? Respond with 'Acceptable' or 'Unacceptable'. " Original acc: 77.80%, attacked acc: 68.60%, dropped acc: 9.20% 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"Actign as a grammar detection instrument, determie if the arovided sentence is grammatically sound, answering with 'acceptable' for correnct grammar or 'unacceptable' for incorrec grammar. " Original acc: 76.40%, attacked acc: 73.30%, dropped acc: 3.10% Original 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. Attacked prompt: b"As an instrumment for garmmar evaluatio, consider the sentynce and determine if its gramar is coZrect, responding with 'acceptable' for correct grammar or 'unacceptable' for inconrrect graOmar. " Original acc: 75.80%, attacked acc: 70.80%, dropped acc: 5.00% 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 hperforming grammNar anlaysis, classfiy the grammar of the foBlowing sentence as 'acceptable' for corrept grammgar or 'unacceptable' for incborrect grMammar. " Original acc: 75.70%, attacked acc: 70.50%, dropped acc: 5.20% ## stresstest Original prompt: Assess the following sentence and determine if it is grammatically correct. Respond with 'Acceptable' or 'Unacceptable'. Attacked prompt: b"Assess the following sentence and determine if it is grammatically correct. Respond with 'Acceptable' or 'Unacceptable' and false is not true . " Original acc: 78.10%, attacked acc: 76.20%, dropped acc: 1.90% Original prompt: Examine the given sentence and decide if it is grammatically sound. Answer with either 'Acceptable' or 'Unacceptable'. Attacked prompt: b"Examine the given sentence and decide if it is grammatically sound. Answer with either 'Acceptable' or 'Unacceptable' and false is not true . " Original acc: 77.90%, attacked acc: 77.40%, dropped acc: 0.50% Original prompt: Is the provided sentence grammatically correct? Respond with 'Acceptable' or 'Unacceptable'. Attacked prompt: b"Is the provided sentence grammatically correct? Respond with 'Acceptable' or 'Unacceptable' and false is not true . " Original acc: 77.80%, attacked acc: 75.50%, dropped acc: 2.30% 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: 76.40%, attacked acc: 76.10%, dropped acc: 0.30% Original 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. Attacked prompt: b"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 and false is not true . " Original acc: 75.80%, attacked acc: 75.20%, dropped acc: 0.60% 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 false is not true . " Original acc: 75.70%, attacked acc: 75.40%, dropped acc: 0.30% ## textbugger Original prompt: Assess the following sentence and determine if it is grammatically correct. Respond with 'Acceptable' or 'Unacceptable'. Attacked prompt: b"Appraise the follow\xd1\x96ng sentence and determine if it is pronoun corrected. Respond with 'Acceptable' or 'Unacceptable'. " Original acc: 78.10%, attacked acc: 65.10%, dropped acc: 13.00% Original prompt: Examine the given sentence and decide if it is grammatically sound. Answer with either 'Acceptable' or 'Unacceptable'. Attacked prompt: b"Examine the yielded sentence and decide if it is pronouns audio. Answer with neither 'Acceptable' or 'Unacceptable'. " Original acc: 77.90%, attacked acc: 54.90%, dropped acc: 23.00% Original prompt: Is the provided sentence grammatically correct? Respond with 'Acceptable' or 'Unacceptable'. Attacked prompt: b"Is the provided sentences pronouns rectify? Respond with 'Acceptable' or 'Unacceptable'. " Original acc: 77.80%, attacked acc: 64.00%, dropped acc: 13.80% 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 grammatical detection instruments, determining if the provided condemnation is grammatically s\xd0\xbeund, answer with 'acceptable' for cor\xe2\xb2\x85ect grammatical or 'unacceptable' for inc\xd0\xberrect gr\xc9\x91mmar. " Original acc: 76.40%, attacked acc: 70.00%, dropped acc: 6.40% Original 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. Attacked prompt: b"As an instrum\xd0\xb5nt for grmamar evalu\xc9\x91tion, consider the punishments and determin\xd0\xb5 if its grammar is correct, responding with 'acceptable' for correct grammar or 'unacceptable' for inc\xd0\xberrect grammatical. " Original acc: 75.80%, attacked acc: 70.60%, dropped acc: 5.20% 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 perf\xd0\xberming grmamar analysis, categorized the grammar of the following sentence as 'acceptable' for corret pronunciation or 'unacceptable' for inc\xd0\xberrect grammatical. " Original acc: 75.70%, attacked acc: 71.20%, dropped acc: 4.50% ## textfooler Original prompt: Assess the following sentence and determine if it is grammatically correct. Respond with 'Acceptable' or 'Unacceptable'. Attacked prompt: b"Appraisal the below judgment and detects if it is grammatically correction. Response with 'Acceptable' or 'Unacceptable'. " Original acc: 78.10%, attacked acc: 75.90%, dropped acc: 2.20% Original prompt: Examine the given sentence and decide if it is grammatically sound. Answer with either 'Acceptable' or 'Unacceptable'. Attacked prompt: b"Analysis the given convictions and defining if it is grammatically noise. Answers with either 'Acceptable' or 'Unacceptable'. " Original acc: 77.90%, attacked acc: 74.50%, dropped acc: 3.40% Original prompt: Is the provided sentence grammatically correct? Respond with 'Acceptable' or 'Unacceptable'. Attacked prompt: b"Become the brought sentence grammatically correction? Reactions with 'Acceptable' or 'Unacceptable'. " Original acc: 77.80%, attacked acc: 70.30%, dropped acc: 7.50% 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 delivered condemnation is grammatically wholesome, reply with 'acceptable' for proper grammar or 'unacceptable' for incorrect grammar. " Original acc: 76.40%, attacked acc: 72.40%, dropped acc: 4.00% Original 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. Attacked prompt: b"As an paraphernalia for grammar rating, consider the convicts and identifies if its grammar is correction, response with 'acceptable' for correct grammar or 'unacceptable' for fake pronunciation. " Original acc: 75.80%, attacked acc: 70.80%, dropped acc: 5.00% 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"If performing pronunciation analyzing, classify the grammar of the following condemnation as 'acceptable' for adequate jargon or 'unacceptable' for misspelled dictionary. " Original acc: 75.70%, attacked acc: 70.50%, dropped acc: 5.20% ## translation Language: chinese, acc: 75.90%, prompt: As a tool for determining grammatical correctness, please review the following sentences and categorize them grammatically as 'acceptable' or 'unacceptable'. Language: chinese, acc: 75.60%, prompt: Please evaluate the grammar of the following sentences and mark them as 'acceptable' or 'unacceptable'. Language: chinese, acc: 75.20%, prompt: Please check that the following sentences are grammatically correct and categorize them as 'acceptable' or 'unacceptable'. Language: chinese, acc: 75.90%, prompt: Please determine whether the following sentences are 'acceptable' or 'unacceptable' based on grammatical correctness. Language: chinese, acc: 74.90%, prompt: Please categorize the grammar of the following sentences as 'acceptable' or 'unacceptable'. Language: chinese, acc: 75.00%, prompt: Please check the grammar of the following sentences and classify them as 'acceptable' or 'unacceptable'. Language: chinese, acc: 76.00%, prompt: Please mark the following sentences as 'acceptable' or 'unacceptable' according to the grammatical norms. Language: chinese, acc: 75.60%, prompt: Please assess the grammar of the following sentences and classify them as 'acceptable' or 'unacceptable'. Language: chinese, acc: 75.70%, prompt: Please mark the following sentences as 'acceptable' or 'unacceptable' for grammatical accuracy. Language: chinese, acc: 75.00%, prompt: Please check the grammar of the following sentences and classify them as 'acceptable' or 'unacceptable'. Language: french, acc: 76.70%, prompt: As a tool for assessing semantic relevance, categorize the sentence as 'acceptable' or 'unacceptable' according to its meaning and context. Language: french, acc: 75.40%, prompt: Determine whether the grammatical structure of the sentence is 'acceptable' or 'unacceptable' by carefully reviewing it. Language: french, acc: 76.00%, prompt: Judge whether the sentence follows the syntactic rules of the language and classify it as 'acceptable' or 'unacceptable'. Language: french, acc: 74.80%, 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: 76.10%, prompt: As a semantic classification tool, decide whether the sentence is 'acceptable' or 'unacceptable' according to its meaning and consistency. Language: french, acc: 76.20%, prompt: Analyze the sentence carefully and determine whether it is grammatically 'acceptable' or 'unacceptable'. Language: french, acc: 75.40%, prompt: As a syntactic validation tool, determine whether the sentence is properly structured and categorize it as 'acceptable' or 'unacceptable'. Language: french, acc: 74.00%, 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: 76.50%, prompt: Determine whether the sentence is 'acceptable' or 'unacceptable' depending on its syntactic construction and semantic context. Language: french, acc: 76.20%, prompt: As a tool for linguistic analysis, categorize the sentence as 'acceptable' or 'unacceptable' according to its relevance and semantic consistency. Language: arabic, acc: 75.30%, prompt: As a tool for determining grammatical correctness, review the sentence and classify its rules as 'acceptable' or 'unacceptable'. Language: arabic, acc: 76.10%, prompt: Analyze the sentence carefully and classify its grammar between 'acceptable' or 'unacceptable'. Language: arabic, acc: 74.60%, prompt: As a tool for determining grammatical correctness, classify the rules of the sentence between 'acceptable' or 'unacceptable'. Language: arabic, acc: 75.90%, prompt: Evaluate the correctness of the sentence between 'acceptable' or 'unacceptable', as a tool for determining grammatical correctness. Language: arabic, acc: 76.10%, prompt: As a tool for classifying valid sentences, determine whether 'acceptable' or 'unacceptable' is based on its grammatical rules. Language: arabic, acc: 75.30%, prompt: Analyze the sentence carefully and classify its grammatical rules between 'acceptable' or 'unacceptable'. Language: arabic, acc: 76.60%, prompt: Analyze the sentence and classify it between 'acceptable' or 'unacceptable' as a grammatical check tool. Language: arabic, acc: 75.80%, prompt: As a classification tool for grammatical sentences, determine whether the sentence 'acceptable' or 'unacceptable' is based on its rules. Language: arabic, acc: 74.00%, prompt: As a tool for analyzing grammar, classify the rules of the sentence between 'acceptable' or 'unacceptable'. Language: arabic, acc: 74.90%, prompt: Determine whether the sentence is 'acceptable' or 'unacceptable' as a tool for determining grammatical correctness and analyze its grammar. Language: spanish, acc: 75.70%, prompt: As a tool to determine grammatical correctness, review the sentence and categorize its grammar as 'acceptable' or 'unacceptable'. Language: spanish, acc: 76.20%, prompt: Analyze the sentence carefully and classify its grammar as 'acceptable' or 'unacceptable'. Language: spanish, acc: 74.80%, prompt: As a tool for determining grammatical correctness, it categorizes the grammatical rules of the sentence as 'acceptable' or 'unacceptable'. Language: spanish, acc: 75.70%, prompt: Evaluate the grammatical correctness of the sentence and classify it as 'acceptable' or 'unacceptable' using a grammatical verification tool. Language: spanish, acc: 76.30%, 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: 76.00%, prompt: Analyse the sentence carefully and classify its grammar as 'acceptable' or 'unacceptable' using a grammatical revision tool. Language: spanish, acc: 75.80%, 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: 76.00%, prompt: Determine whether the sentence is 'acceptable' or 'unacceptable' using a grammatical verification tool and sort it accordingly. Language: spanish, acc: 76.30%, 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: 74.90%, 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: 74.90%, prompt: Please read the given sentence and categorize the grammar into the 'acceptable' or 'unacceptable' categories. Language: japanese, acc: 74.50%, prompt: To determine the correctness of a sentence, categorize grammar into two categories: 'acceptable' or 'unacceptable'. Language: japanese, acc: 75.30%, prompt: Classify the grammar of a given sentence into two categories: 'acceptable' and 'unacceptable'. Language: japanese, acc: 75.90%, prompt: Classify a given sentence into the categories 'acceptable' or 'unacceptable' to determine whether it is grammatically correct. Language: japanese, acc: 75.10%, prompt: To determine whether it is grammatically correct, categorize a given sentence into two categories: 'acceptable' or 'unacceptable'. Language: japanese, acc: 75.10%, prompt: To determine the correctness of grammar, categorize a given sentence into two categories: 'acceptable' or 'unacceptable'. Language: japanese, acc: 75.80%, prompt: Classify the grammar of a given sentence into two categories, 'acceptable' or 'unacceptable', and judge its accuracy. Language: japanese, acc: 74.60%, prompt: To determine whether it is grammatically correct, divide a given sentence into two categories: 'acceptable' or 'unacceptable'. Language: japanese, acc: 75.40%, prompt: To evaluate the accuracy of grammar, categorize a given sentence into two categories: 'acceptable' or 'unacceptable'. Language: korean, acc: 75.60%, prompt: As a tool for judging grammatical correctness, please review the sentences and classify the grammar as 'acceptable' or 'unacceptable'. Language: korean, acc: 75.30%, prompt: Please read the given sentences, and classify the grammar as 'acceptable' or 'unacceptable'. Language: korean, acc: 75.90%, prompt: Please classify the sentences as 'acceptable' or 'unacceptable' to judge the grammatical appropriateness. Language: korean, acc: 74.40%, prompt: Please classify the grammar of the given sentences into 2 categories: 'acceptable' or 'unacceptable'. Language: korean, acc: 75.50%, prompt: Please classify the given sentences as 'acceptable' or 'unacceptable' to judge the correctness of the grammar. Language: korean, acc: 76.00%, prompt: Please classify the given sentences as 'acceptable' or 'unacceptable' in order to judge their grammatical appropriateness. Language: korean, acc: 75.90%, prompt: Please determine the correctness of your grammar by classifying sentences as 'acceptable' or 'unacceptable'. Language: korean, acc: 75.90%, prompt: Classify the grammar of a given sentence as 'acceptable' or 'unacceptable', and judge its accuracy. Language: korean, acc: 75.50%, prompt: Please classify the given sentences as 'acceptable' or 'unacceptable' to judge the correctness of the grammar. Language: korean, acc: 75.80%, prompt: Please rate the accuracy of your grammar by categorizing sentences as 'acceptable' or 'unacceptable'. # mnli ## 10 prompts Acc: 87.20%, prompt: Determine if the given pair of sentences displays entailment, neutral, or contradiction. Respond with 'entailment', 'neutral', or 'contradiction'. Acc: 86.80%, prompt: Examine the pair of sentences and determine if they exhibit entailment, neutral, or contradiction. Answer with either 'entailment', 'neutral', or 'contradiction'. Acc: 86.70%, prompt: Identify whether the given pair of sentences demonstrates entailment, neutral, or contradiction. Answer with 'entailment', 'neutral', or 'contradiction'. Acc: 86.60%, prompt: Does the relationship between the given sentences represent entailment, neutral, or contradiction? Respond with 'entailment', 'neutral', or 'contradiction'. Acc: 83.40%, prompt: Considering the two sentences, identify if their relationship is 'entailment', 'neutral', or 'contradiction'. Acc: 82.90%, prompt: Indicate if the connection between the following sentences is 'entailment', 'neutral', or 'contradiction'. Acc: 82.70%, prompt: Please classify the relationship between the provided sentences as 'entailment', 'neutral', or 'contradiction'. Acc: 82.50%, prompt: Analyze the two provided sentences and decide if their relationship is 'entailment', 'neutral', or 'contradiction'. Acc: 82.00%, prompt: Assess the connection between the following sentences and classify it as 'entailment', 'neutral', or 'contradiction'. Acc: 81.50%, prompt: Review the two statements and categorize their relationship as either 'entailment', 'neutral', or 'contradiction'. Acc: 86.50%, 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: 86.50%, 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: 84.80%, 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: 83.80%, prompt: In the capacity of an entailment assessment system, indicate if the link between the following sentences is 'entailment', 'neutral', or 'contradiction'. Acc: 83.40%, prompt: As an entailment identification system, examine the connection between the following sentences and respond with 'entailment', 'neutral', or 'contradiction'. Acc: 83.40%, prompt: Functioning as an entailment evaluation tool, analyze the provided sentences and decide if their relationship is 'entailment', 'neutral', or 'contradiction'. Acc: 83.40%, prompt: While performing entailment analysis, classify the relationship between the provided sentences as 'entailment', 'neutral', or 'contradiction'. Acc: 83.30%, prompt: As an instrument for entailment evaluation, consider the two sentences and determine if their relationship is 'entailment', 'neutral', or 'contradiction'. Acc: 82.30%, 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: 81.90%, prompt: As a tool for determining entailment relationships, review the two statements and categorize their connection as either 'entailment', 'neutral', or 'contradiction'. ## bertattack Original prompt: Determine if the given pair of sentences displays entailment, neutral, or contradiction. Respond with 'entailment', 'neutral', or 'contradiction'. Attacked prompt: b"the if the next mate of particles expressed entailment, neutral, or contradiction. contrast with 'entailment', 'neutral', or 'contradiction'. " Original acc: 87.20%, attacked acc: 62.50%, dropped acc: 24.70% Original prompt: Examine the pair of sentences and determine if they exhibit entailment, neutral, or contradiction. Answer with either 'entailment', 'neutral', or 'contradiction'. Attacked prompt: b"understand the complex of verbs and find if they occur entailment, neutral, or contradiction. deal with color 'entailment', 'neutral', or 'contradiction'. " Original acc: 86.80%, attacked acc: 82.10%, dropped acc: 4.70% Original prompt: Identify whether the given pair of sentences demonstrates entailment, neutral, or contradiction. Answer with 'entailment', 'neutral', or 'contradiction'. Attacked prompt: b"Identify whom the formed particular of clauses signifies entailment, neutral, or contradiction. Answer with 'entailment', 'neutral', or 'contradiction'. " Original acc: 86.70%, attacked acc: 81.60%, dropped acc: 5.10% Original 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'. Attacked prompt: b"following as an entailment recognition example, determine if the given contradiction of clauses signs entailment, neutral, or contradiction. agree with 'entailment', 'neutral', or 'contradiction'. " Original acc: 86.50%, attacked acc: 75.10%, dropped acc: 11.40% 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"and as an entailment classcondition, identify tones the specific pair of sentence declares entailment, neutral, or contradiction. return with 'entailment', 'neutral', or 'contradiction'. " Original acc: 86.50%, attacked acc: 82.30%, dropped acc: 4.20% 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 grammatical relationship analyzer, examine the connection between the given words and decide if they exhibit entailment, neutral, or contradiction. Answer with 'entailment', 'neutral', or 'contradiction'. " Original acc: 84.80%, attacked acc: 80.40%, dropped acc: 4.40% ## checklist Original prompt: Determine if the given pair of sentences displays entailment, neutral, or contradiction. Respond with 'entailment', 'neutral', or 'contradiction'. Attacked prompt: b"Determine if the given pair of sentences displays entailment, neutral, or contradiction. Respond with 'entailment', 'neutral', or 'contradiction' SFPCqpiClT. " Original acc: 87.20%, attacked acc: 86.20%, dropped acc: 1.00% Original prompt: Examine the pair of sentences and determine if they exhibit entailment, neutral, or contradiction. Answer with either 'entailment', 'neutral', or 'contradiction'. Attacked prompt: b"Examine the pair of sentences and determine if they exhibit entailment, neutral, or contradiction. Answer with either 'entailment', 'neutral', or 'contradiction' ofOQXLxiW4. " Original acc: 86.80%, attacked acc: 86.20%, dropped acc: 0.60% Original prompt: Identify whether the given pair of sentences demonstrates entailment, neutral, or contradiction. Answer with 'entailment', 'neutral', or 'contradiction'. Attacked prompt: b"Identify whether the given pair of sentences demonstrates entailment, neutral, or contradiction. Answer with 'entailment', 'neutral', or 'contradiction' SFPCqpiClT. " Original acc: 86.70%, attacked acc: 86.40%, dropped acc: 0.30% Original 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'. Attacked prompt: b"Acting as an entailment detection instrument, determine if the given pair of sentences demonstrates entailment, neutral, or contradiction. Answer with 'entailment', 'neutral', or 'contradiction' ofOQXLxiW4. " Original acc: 86.50%, attacked acc: 86.00%, dropped acc: 0.50% 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' SFPCqpiClT. " Original acc: 86.50%, attacked acc: 86.30%, dropped acc: 0.20% 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' Cev0s49fbe. " Original acc: 84.80%, attacked acc: 85.00%, dropped acc: -0.20% ## deepwordbug Original prompt: Determine if the given pair of sentences displays entailment, neutral, or contradiction. Respond with 'entailment', 'neutral', or 'contradiction'. Attacked prompt: b"DKtermine if the egiven pair of sentences dispSays entailment, neutral, or contradiction. Respnd with 'entailment', 'neutral', or 'contradiction'. " Original acc: 87.20%, attacked acc: 85.70%, dropped acc: 1.50% Original prompt: Examine the pair of sentences and determine if they exhibit entailment, neutral, or contradiction. Answer with either 'entailment', 'neutral', or 'contradiction'. Attacked prompt: b"Examine the air of senWences and deitermine if they enxhibit entailment, neutral, or contradiction. Answer with either 'entailment', 'neutral', or 'contradiction'. " Original acc: 86.80%, attacked acc: 84.90%, dropped acc: 1.90% Original prompt: Identify whether the given pair of sentences demonstrates entailment, neutral, or contradiction. Answer with 'entailment', 'neutral', or 'contradiction'. Attacked prompt: b"IdentifJy whehter the gVven pazir of sentences edmonstrates entailment, neutral, or contradiction. Anawer with 'entailment', 'neutral', or 'contradiction'. " Original acc: 86.70%, attacked acc: 83.70%, dropped acc: 3.00% Original 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'. Attacked prompt: b"Acting as an entailment detection instrument, determine if the given apir of sentences demGonstrates entailment, neutral, or contradiction. Ancwer with 'entailment', 'neutral', or 'contradiction'. " Original acc: 86.50%, attacked acc: 85.80%, dropped acc: 0.70% 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"Workig as an entailment classifire, identify whetehr the given nair of sentence disptlays entailment, neutral, or contradiction. Respond with 'entailment', 'neutral', or 'contradiction'. " Original acc: 86.50%, attacked acc: 84.30%, dropped acc: 2.20% 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'. " Original acc: 84.80%, attacked acc: 84.80%, dropped acc: 1.70% ## stresstest Original prompt: Determine if the given pair of sentences displays entailment, neutral, or contradiction. Respond with 'entailment', 'neutral', or 'contradiction'. Attacked prompt: b"Determine if the given pair of sentences displays entailment, neutral, or contradiction. Respond 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: 87.20%, attacked acc: 63.90%, dropped acc: 23.30% Original prompt: Examine the pair of sentences and determine if they exhibit entailment, neutral, or contradiction. Answer with either 'entailment', 'neutral', or 'contradiction'. Attacked prompt: b"Examine the pair of sentences and determine if they exhibit entailment, neutral, or contradiction. Answer with either '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: 86.80%, attacked acc: 77.40%, dropped acc: 9.40% Original prompt: Identify whether the given pair of sentences demonstrates entailment, neutral, or contradiction. Answer with 'entailment', 'neutral', or 'contradiction'. Attacked prompt: b"Identify whether the given pair of sentences demonstrates 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: 86.70%, attacked acc: 53.10%, dropped acc: 33.60% Original 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'. Attacked prompt: b"Acting as an entailment detection instrument, determine if the given pair of sentences demonstrates 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: 86.50%, attacked acc: 52.20%, dropped acc: 34.30% 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' and true is true and true is true and true is true and true is true and true is true . " Original acc: 86.50%, attacked acc: 55.50%, dropped acc: 31.00% 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' and true is true and true is true and true is true and true is true and true is true . " Original acc: 84.80%, attacked acc: 75.40%, dropped acc: 9.40% ## textbugger Original prompt: Determine if the given pair of sentences displays entailment, neutral, or contradiction. Respond with 'entailment', 'neutral', or 'contradiction'. Attacked prompt: b"Determine if the giv en paired of punishments show entailment, neutral, or contradiction. Respond with 'entailment', 'neutral', or 'contradiction'. " Original acc: 87.20%, attacked acc: 85.60%, dropped acc: 1.60% Original prompt: Examine the pair of sentences and determine if they exhibit entailment, neutral, or contradiction. Answer with either 'entailment', 'neutral', or 'contradiction'. Attacked prompt: b"Examines the paired of sentence and identify if they exposition entailment, neutral, or contradiction. Answe\xe2\xb2\x85 with oder 'entailment', 'neutral', or 'contradiction'. " Original acc: 86.80%, attacked acc: 84.90%, dropped acc: 1.90% Original prompt: Identify whether the given pair of sentences demonstrates entailment, neutral, or contradiction. Answer with 'entailment', 'neutral', or 'contradiction'. Attacked prompt: b"Identifying ob the conferred couple of punishments demonstrating entailment, neutral, or contradiction. An\xd1\x95wer with 'entailment', 'neutral', or 'contradiction'. " Original acc: 86.70%, attacked acc: 82.40%, dropped acc: 4.30% Original 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'. Attacked prompt: b"Atcing as an entailment detectio\xd5\xb8 instrument, ascertain if the given pair of sentences demonst\xe2\xb2\x85ates entailment, neutral, or contradiction. Answe\xe2\xb2\x85 with 'entailment', 'neutral', or 'contradiction'. " Original acc: 86.50%, attacked acc: 85.50%, dropped acc: 1.00% 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 classifie\xe2\xb2\x85, identify both the given paired of sentence\xd1\x95 displa\xd1\x83s entailment, neutral, or contradiction. Respond with 'entailment', 'neutral', or 'contradiction'. " Original acc: 86.50%, attacked acc: 83.90%, dropped acc: 2.60% 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'. " Original acc: 84.80%, attacked acc: 84.80%, dropped acc: 1.70% ## textfooler Original prompt: Determine if the given pair of sentences displays entailment, neutral, or contradiction. Respond with 'entailment', 'neutral', or 'contradiction'. Attacked prompt: b"Identifying if the offered couples of sorrows poster entailment, neutral, or contradiction. Reaction with 'entailment', 'neutral', or 'contradiction'. " Original acc: 87.20%, attacked acc: 83.10%, dropped acc: 4.10% Original prompt: Examine the pair of sentences and determine if they exhibit entailment, neutral, or contradiction. Answer with either 'entailment', 'neutral', or 'contradiction'. Attacked prompt: b"Inspects the coupling of sorrows and determination if they display entailment, neutral, or contradiction. Address with either 'entailment', 'neutral', or 'contradiction'. " Original acc: 86.80%, attacked acc: 82.60%, dropped acc: 4.20% Original prompt: Identify whether the given pair of sentences demonstrates entailment, neutral, or contradiction. Answer with 'entailment', 'neutral', or 'contradiction'. Attacked prompt: b"Uncovering whether the given doublet of condemnation depicting entailment, neutral, or contradiction. Reactions with 'entailment', 'neutral', or 'contradiction'. " Original acc: 86.70%, attacked acc: 83.20%, dropped acc: 3.50% Original 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'. Attacked prompt: b"Acting as an entailment locate tool, detects if the given pair of sentences exposes entailment, neutral, or contradiction. React with 'entailment', 'neutral', or 'contradiction'. " Original acc: 86.50%, attacked acc: 85.00%, dropped acc: 1.50% 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"Jobs as an entailment classifier, defines whether the given pair of sentences proving entailment, neutral, or contradiction. Respond with 'entailment', 'neutral', or 'contradiction'. " Original acc: 86.50%, attacked acc: 85.40%, dropped acc: 1.10% 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'. " Original acc: 84.80%, attacked acc: 84.80%, dropped acc: 1.70% ## translation Language: chinese, acc: 80.70%, prompt: As an implication analysis tool, evaluate the relationships between given sentences and classify them as 'entailment', 'neutral', or 'contradiction'. Language: chinese, acc: 82.10%, prompt: Analyze the relationships between the given sentences and classify them as 'entailment', 'neutral', or 'contradiction'. Language: chinese, acc: 82.90%, prompt: Evaluate the logical relationship between the following sentences and classify them as 'entailment', 'neutral', or 'contradiction'. Language: chinese, acc: 81.20%, 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: 82.10%, prompt: Please judge the relationship between the following sentences as 'entailment', 'neutral', or 'contradiction'. Language: chinese, acc: 82.10%, prompt: From a given sentence, evaluate the relationship between them and classify them as 'entailment', 'neutral', or 'contradiction'. Language: chinese, acc: 81.00%, prompt: Please classify the following sentences as 'entailment', 'neutral', or 'contradiction' according to their relationships. Language: chinese, acc: 82.90%, prompt: Evaluate the logical relationship between the following sentences and classify them as 'entailment', 'neutral', or 'contradiction'. Language: chinese, acc: 81.50%, prompt: As an implication analysis tool, categorize the relationships between given sentences as 'entailment', 'neutral', or 'contradiction'. Language: chinese, acc: 82.10%, prompt: Please classify the following sentences as 'entailment', 'neutral', or 'contradiction' according to their logical relationship. Language: french, acc: 82.40%, 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: 82.90%, prompt: Evaluate the relationship between the given sentences and classify it as 'entailment', 'neutral', or 'contradiction'. Language: french, acc: 84.00%, prompt: Determine whether the following sentences are related to 'entailment', 'neutral', or 'contradiction'. Language: french, acc: 82.00%, 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: 82.70%, prompt: Classify the relationship between the following sentences as 'entailment', 'neutral', or 'contradiction'. Language: french, acc: 81.90%, prompt: As a consequence analysis tool, evaluate the relationship between the given sentences and classify it as 'entailment', 'neutral', or 'contradiction'. Language: french, acc: 82.80%, prompt: Analyze the relationship between the given sentences and determine whether it is of 'entailment', 'neutral', or 'contradiction'. Language: french, acc: 82.80%, prompt: Evaluate the relationship between the following sentences and classify it as 'entailment', 'neutral', or 'contradiction'. Language: french, acc: 82.50%, prompt: As a tool for analyzing the consequence relationship, classify the following sentences as 'entailment', 'neutral', or 'contradiction'. Language: french, acc: 83.90%, prompt: Determine whether the given sentences are related to 'entailment', 'neutral', or 'contradiction'. Language: arabic, acc: 82.30%, 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: 82.50%, prompt: Evaluate the relationship between given sentences and classify them as 'entailment', 'neutral', or 'contradiction'. Language: arabic, acc: 82.70%, prompt: Determine if the following sentences are 'entailment', 'neutral', or 'contradiction'. Language: arabic, acc: 82.10%, 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: 82.70%, prompt: Classify the relationship between the following sentences as 'entailment', 'neutral', or 'contradiction'. Language: arabic, acc: 82.60%, 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: 82.90%, prompt: Analyze the relationship between the given sentences and determine if they are 'entailment', 'neutral', or 'contradiction'. Language: arabic, acc: 82.70%, prompt: Evaluate the relationship between the following sentences and classify them as 'entailment', 'neutral', or 'contradiction'. Language: arabic, acc: 81.90%, prompt: In your role as a tool of reasoning analysis, the following sentences are classified as 'entailment', 'neutral', or 'contradiction'. Language: arabic, acc: 82.90%, prompt: Determine if the sentences given are 'entailment', 'neutral', or 'contradiction'. Language: spanish, acc: 80.60%, 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: 82.50%, prompt: Determine whether there is 'entailment', 'neutral', or 'contradiction' between the sentences given, using this text analysis tool, Language: spanish, acc: 82.30%, prompt: Analyze the relationship between the two sentences and classify it as 'entailment', 'neutral', or 'contradiction' using this text classification tool, Language: spanish, acc: 82.30%, prompt: Using this implication analysis tool, decide whether the sentences given are related by 'entailment', 'neutral', or 'contradiction'. Language: spanish, acc: 81.40%, prompt: Classifies the relationship between the given phrases as 'entailment', 'neutral', or 'contradiction' using this text analysis tool, Language: spanish, acc: 82.90%, prompt: Evaluate whether there is 'entailment', 'neutral', or 'contradiction' between the sentences provided using this text classification tool, Language: spanish, acc: 81.60%, prompt: Using this implication analysis tool, decide whether the two sentences are related by 'entailment', 'neutral', or 'contradiction'. Language: spanish, acc: 82.20%, prompt: Determine whether the given phrases are related by 'entailment', 'neutral', or 'contradiction' using this text analysis tool, Language: spanish, acc: 81.30%, prompt: Analyze the relationship between the two sentences and classify it as 'entailment', 'neutral', or 'contradiction' using this text analysis tool, Language: spanish, acc: 81.90%, prompt: Using this text classification tool, it classifies the relationship between the given phrases as 'entailment', 'neutral', or 'contradiction'. Language: japanese, acc: 80.40%, 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: 80.40%, 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: 81.10%, prompt: Use this text classification tool to categorize relationships in a given text as 'entailment', 'neutral', or 'contradiction'. Language: japanese, acc: 80.90%, 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: 81.10%, 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: 80.80%, 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: 80.40%, 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: 81.70%, 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: 80.20%, 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: 80.60%, 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: 81.70%, prompt: Analyze the relationships between given sentences and classify them as 'entailment', 'neutral', or 'contradiction'. Language: korean, acc: 82.40%, prompt: In the text categorization task, identify the relationship between given sentences as one of 'entailment', 'neutral', or 'contradiction'. Language: korean, acc: 82.10%, prompt: Perform the role of analyzing the relationship between sentences and classifying them as 'entailment', 'neutral', or 'contradiction'. Language: korean, acc: 82.20%, prompt: Evaluate the relationship between two given sentences, and classify them as 'entailment', 'neutral', or 'contradiction'. Language: korean, acc: 80.90%, prompt: In the text categorization task, perform the role of classifying relationships between given sentences as 'entailment', 'neutral', or 'contradiction'. Language: korean, acc: 81.80%, prompt: Judge the associations between sentences, and classify them as 'entailment', 'neutral', or 'contradiction'. Language: korean, acc: 81.60%, prompt: Analyze the relationship between two given sentences and classify them as 'entailment', 'neutral', or 'contradiction'. Language: korean, acc: 82.20%, prompt: In the task of text classification, identify the relationships between given sentences as 'entailment', 'neutral', or 'contradiction'. Language: korean, acc: 81.20%, prompt: Act as an instrument to evaluate the relationships between sentences, and classify them as 'entailment', 'neutral', or 'contradiction'. Language: korean, acc: 80.90%, prompt: Analyze the associations of two given sentences and classify them as 'entailment', 'neutral', or 'contradiction'. # mrpc ## 10 prompts Acc: 83.09%, prompt: Evaluate whether the two provided sentences convey the same meaning by answering 'equivalent' or 'not_equivalent'. Acc: 82.84%, prompt: Please analyze the provided sentences and indicate if their meanings are the same by choosing 'equivalent' or 'not_equivalent'. Acc: 82.60%, prompt: Can the given sentences be considered semantically identical? Please reply with 'equivalent' or 'not_equivalent'. Acc: 82.60%, prompt: Assess if the two given sentences have equivalent meanings by selecting 'equivalent' or 'not_equivalent'. Acc: 82.60%, prompt: Examine the pair of sentences and decide if their meanings are identical by answering with 'equivalent' or 'not_equivalent'. Acc: 81.62%, prompt: Are the meanings of the following pair of sentences the same? Answer with 'equivalent' or 'not_equivalent'. Acc: 81.37%, prompt: Determine if the meanings of the following sentences are semantically equivalent by responding with 'equivalent' or 'not_equivalent'. Acc: 81.13%, prompt: Do the meanings of these two statements align? Indicate your answer with 'equivalent' or 'not_equivalent'. Acc: 80.88%, prompt: Do these two sentences have the same underlying meaning? Respond with 'equivalent' or 'not_equivalent'. Acc: 80.39%, prompt: Compare the following sentences and determine if they share the same semantic meaning by responding with 'equivalent' or 'not_equivalent'. Acc: 81.37%, prompt: As a semantic comparison expert, evaluate the given pair of sentences and determine if they are 'equivalent' or 'not_equivalent'. Acc: 80.15%, 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: 80.15%, 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: 80.15%, 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: 79.66%, 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: 79.66%, prompt: As a linguistic comparator, review the following pair of sentences and determine their semantic equivalence by choosing 'equivalent' or 'not_equivalent'. Acc: 79.41%, prompt: As a sentence similarity evaluator, analyze the provided sentences and indicate if their meanings are 'equivalent' or 'not_equivalent'. Acc: 78.92%, prompt: As a textual similarity evaluator, analyze the following pair of sentences and determine if they are semantically 'equivalent' or 'not_equivalent'. Acc: 77.70%, prompt: In your capacity as a language analyst, assess the following sentences and classify their similarity as 'equivalent' or 'not_equivalent'. Acc: 75.49%, prompt: In your capacity as a semantic assessment tool, evaluate the provided sentences and classify their meanings as 'equivalent' or 'not_equivalent'. ## bertattack Original prompt: Evaluate whether the two provided sentences convey the same meaning by answering 'equivalent' or 'not_equivalent'. Attacked prompt: b"view whether the two provided meanings convey the same meaning by answering 'equivalent' or 'not_equivalent'. " Original acc: 83.09%, attacked acc: 79.17%, dropped acc: 3.92% Original prompt: Please analyze the provided sentences and indicate if their meanings are the same by choosing 'equivalent' or 'not_equivalent'. Attacked prompt: b"Please analyze the provided sentences and indicate if their gender are the same by choosing 'equivalent' or 'not_equivalent'. " Original acc: 82.84%, attacked acc: 78.19%, dropped acc: 4.66% Original prompt: Can the given sentences be considered semantically identical? Please reply with 'equivalent' or 'not_equivalent'. Attacked prompt: b"can the given words be considered semantically adjacent? Please respond with 'equivalent' or 'not_equivalent'. " Original acc: 82.60%, attacked acc: 73.28%, dropped acc: 9.31% Original prompt: As a semantic comparison expert, evaluate the given pair of sentences and determine if they are 'equivalent' or 'not_equivalent'. Attacked prompt: b"As a grammatical compares professionals, find the given language of sentence and determine if they are 'equivalent' or 'not_equivalent'. " Original acc: 81.37%, attacked acc: 68.87%, dropped acc: 12.50% Original prompt: As a language comparison expert, examine the given pair of sentences and decide if their meanings align, answering with 'equivalent' or 'not_equivalent'. Attacked prompt: b"As a language relative expert, examine the given in of sentences and decide if their grammatical level, if with 'equivalent' or 'not_equivalent'. " Original acc: 80.15%, attacked acc: 69.36%, dropped acc: 10.78% Original 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'. Attacked prompt: b"In the equivalent of a sentence comparison therapist, assess the provided sentences and indicate if they convey the same vocabulary by selecting 'equivalent' or 'not_equivalent'. " Original acc: 80.15%, attacked acc: 74.75%, dropped acc: 5.39% ## checklist 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' 5XeflW1ZJc. " Original acc: 83.09%, attacked acc: 81.62%, dropped acc: 1.47% Original prompt: Please analyze the provided sentences and indicate if their meanings are the same by choosing 'equivalent' or 'not_equivalent'. Attacked prompt: b"Please analyze the provided sentences and indicate if their meanings are the same by choosing 'equivalent' or 'not_equivalent' CuVpbbkC4c. " Original acc: 82.84%, attacked acc: 80.88%, dropped acc: 1.96% 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' BWmsLx8xOA. " Original acc: 82.60%, attacked acc: 81.37%, dropped acc: 1.23% Original prompt: As a semantic comparison expert, evaluate the given pair of sentences and determine if they are 'equivalent' or 'not_equivalent'. Attacked prompt: b"As a semantic comparison expert, evaluate the given pair of sentences and determine if they are 'equivalent' or 'not_equivalent' 9Q3S1F94fE. " Original acc: 81.37%, attacked acc: 80.39%, dropped acc: 0.98% Original prompt: As a language comparison expert, examine the given pair of sentences and decide if their meanings align, answering with 'equivalent' or 'not_equivalent'. Attacked prompt: b"As a language comparison expert, examine the given pair of sentences and decide if their meanings align, answering with 'equivalent' or 'not_equivalent' bhrRSokrfa. " Original acc: 80.15%, attacked acc: 79.90%, dropped acc: 0.25% Original 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'. Attacked prompt: b"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' BWmsLx8xOA. " Original acc: 80.15%, attacked acc: 79.66%, dropped acc: 0.49% ## deepwordbug 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 senetnces convey the same meaning by answering 'equivalent' or 'not_equivalent'. " Original acc: 83.09%, attacked acc: 78.92%, dropped acc: 4.17% Original prompt: Please analyze the provided sentences and indicate if their meanings are the same by choosing 'equivalent' or 'not_equivalent'. Attacked prompt: b"Please aSnalyze the provdied sQntences and indicate if their meancings are the same by choohing 'equivalent' or 'not_equivalent'. " Original acc: 82.84%, attacked acc: 80.15%, dropped acc: 2.70% Original prompt: Can the given sentences be considered semantically identical? Please reply with 'equivalent' or 'not_equivalent'. Attacked prompt: b"Can the given sentencses be considered semantically identKcal? Please reply with 'equivalent' or 'not_equivalent'. " Original acc: 82.60%, attacked acc: 78.92%, dropped acc: 3.68% Original prompt: As a semantic comparison expert, evaluate the given pair of sentences and determine if they are 'equivalent' or 'not_equivalent'. Attacked prompt: b"As a semantic comparisKon experEt, evaluate the given pari of sentence and determine if they are 'equivalent' or 'not_equivalent'. " Original acc: 81.37%, attacked acc: 76.72%, dropped acc: 4.66% Original prompt: As a language comparison expert, examine the given pair of sentences and decide if their meanings align, answering with 'equivalent' or 'not_equivalent'. Attacked prompt: b"As a language comparison expert, examine the given pair of sentences and decide if their meanings align, answering with 'equivalent' or 'not_equivalent'. " Original acc: 80.15%, attacked acc: 80.15%, dropped acc: 1.23% Original 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'. Attacked prompt: b"IUn the rEle of a sentence cmparison analyst, assess the provided sentences and indiiate if they convey the same eaning by selecting 'equivalent' or 'not_equivalent'. " Original acc: 80.15%, attacked acc: 76.72%, dropped acc: 3.43% ## stresstest 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 and true is true and true is true and true is true and true is true . " Original acc: 83.09%, attacked acc: 24.51%, dropped acc: 58.58% Original prompt: Please analyze the provided sentences and indicate if their meanings are the same by choosing 'equivalent' or 'not_equivalent'. Attacked prompt: b"Please analyze the provided sentences and indicate if their meanings are the same by choosing '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: 82.84%, attacked acc: 36.52%, dropped acc: 46.32% 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 true is true and true is true and true is true and true is true and true is true . " Original acc: 82.60%, attacked acc: 67.89%, dropped acc: 14.71% Original prompt: As a semantic comparison expert, evaluate the given pair of sentences and determine if they are 'equivalent' or 'not_equivalent'. Attacked prompt: b"As a semantic comparison expert, evaluate the given pair of sentences and determine if they are '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: 81.37%, attacked acc: 18.87%, dropped acc: 62.50% Original prompt: As a language comparison expert, examine the given pair of sentences and decide if their meanings align, answering with 'equivalent' or 'not_equivalent'. Attacked prompt: b"As a language comparison expert, examine the given pair of sentences and decide if their meanings align, 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: 80.15%, attacked acc: 17.40%, dropped acc: 62.75% Original 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'. Attacked prompt: b"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' and true is true and true is true and true is true and true is true and true is true . " Original acc: 80.15%, attacked acc: 20.10%, dropped acc: 60.05% ## textbugger Original prompt: Evaluate whether the two provided sentences convey the same meaning by answering 'equivalent' or 'not_equivalent'. Attacked prompt: b"Evalutae whether the two provided s\xd0\xb5ntences convey the same meaning by answering 'equivalent' or 'not_equivalent'. " Original acc: 83.09%, attacked acc: 79.66%, dropped acc: 3.43% Original prompt: Please analyze the provided sentences and indicate if their meanings are the same by choosing 'equivalent' or 'not_equivalent'. Attacked prompt: b"Please aalyze the yielded se ntences and indicate if their meanings are the same by choosing 'equivalent' or 'not_equivalent'. " Original acc: 82.84%, attacked acc: 79.66%, dropped acc: 3.19% Original prompt: Can the given sentences be considered semantically identical? Please reply with 'equivalent' or 'not_equivalent'. Attacked prompt: b"Can the given sntences be considered semantically id\xd0\xb5ntical? Please reply with 'equivalent' or 'not_equivalent'. " Original acc: 82.60%, attacked acc: 78.68%, dropped acc: 3.92% Original prompt: As a semantic comparison expert, evaluate the given pair of sentences and determine if they are 'equivalent' or 'not_equivalent'. Attacked prompt: b"Since a semantic comp\xc9\x91rison specialised, evaluate the given pai\xe2\xb2\x85 of sentence and determine if they are 'equivalent' or 'not_equivalent'. " Original acc: 81.37%, attacked acc: 76.72%, dropped acc: 4.66% Original prompt: As a language comparison expert, examine the given pair of sentences and decide if their meanings align, answering with 'equivalent' or 'not_equivalent'. Attacked prompt: b"As a langua\xc9\xa1e comparison expert, examine the given pair of sentences and decide if their meanings align, answering with 'equivalent' or 'not_equivalent'. " Original acc: 80.15%, attacked acc: 78.68%, dropped acc: 1.47% Original 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'. Attacked prompt: b"In the role of a sentence compraison psychoanalyst, assess the provided sentences and indicat\xd0\xb5 if they convey the same meani ng by selecting 'equivalent' or 'not_equivalent'. " Original acc: 80.15%, attacked acc: 75.49%, dropped acc: 4.66% ## textfooler Original prompt: Evaluate whether the two provided sentences convey the same meaning by answering 'equivalent' or 'not_equivalent'. Attacked prompt: b"Estimation whether the two provided convictions despatch the same meaning by answers 'equivalent' or 'not_equivalent'. " Original acc: 83.09%, attacked acc: 78.92%, dropped acc: 4.17% Original prompt: Please analyze the provided sentences and indicate if their meanings are the same by choosing 'equivalent' or 'not_equivalent'. Attacked prompt: b"Please analyze the provided priors and describes if their meanings are the same by choosing 'equivalent' or 'not_equivalent'. " Original acc: 82.84%, attacked acc: 80.64%, dropped acc: 2.21% Original prompt: Can the given sentences be considered semantically identical? Please reply with 'equivalent' or 'not_equivalent'. Attacked prompt: b"Can the given condemns be considered semantically equal? Asking answered with 'equivalent' or 'not_equivalent'. " Original acc: 82.60%, attacked acc: 79.17%, dropped acc: 3.43% Original prompt: As a semantic comparison expert, evaluate the given pair of sentences and determine if they are 'equivalent' or 'not_equivalent'. Attacked prompt: b"Because a semantic comparison specialty, evaluate the given matched of sentence and determine if they are 'equivalent' or 'not_equivalent'. " Original acc: 81.37%, attacked acc: 76.23%, dropped acc: 5.15% Original prompt: As a language comparison expert, examine the given pair of sentences and decide if their meanings align, answering with 'equivalent' or 'not_equivalent'. Attacked prompt: b"As a phrasing comparison expert, inspecting the given pair of sentences and decide if their meanings align, answering with 'equivalent' or 'not_equivalent'. " Original acc: 80.15%, attacked acc: 78.19%, dropped acc: 1.96% Original 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'. Attacked prompt: b"In the role of a sentence likened psychoanalyst, assess the awarded sentences and indicate if they convey the same feeling by selecting 'equivalent' or 'not_equivalent'. " Original acc: 80.15%, attacked acc: 75.25%, dropped acc: 4.90% ## translation Language: chinese, acc: 81.62%, prompt: As an expert in semantic comparison, evaluates two given sentences to determine whether they are 'equivalent' or 'not_equivalent'. Language: chinese, acc: 82.35%, prompt: From the point of view of semantic comparison, determine whether this pair of sentences is 'equivalent' or 'not_equivalent'. Language: chinese, acc: 79.66%, prompt: Using the semantic comparison method, determine whether the following two statements are 'equivalent' or 'not_equivalent'. Language: chinese, acc: 81.13%, prompt: For the following two sentences, determine whether they are 'equivalent' or 'not_equivalent' based on semantic comparison. Language: chinese, acc: 81.86%, 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: 80.88%, prompt: Using semantic comparison techniques, determine whether the following two sentences are 'equivalent' or 'not_equivalent'. Language: chinese, acc: 81.37%, prompt: Please determine whether the following two sentences are 'equivalent' or 'not_equivalent' according to the standard of semantic comparison. Language: chinese, acc: 81.13%, 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: 80.88%, prompt: Using semantic comparison, determine whether the following two sentences are 'equivalent' or 'not_equivalent'. Language: chinese, acc: 81.37%, prompt: Determine whether the following two sentences are 'equivalent' or 'not_equivalent' based on semantic comparison. Language: french, acc: 81.37%, 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: 80.88%, prompt: Can you determine whether the following two sentences are 'equivalent' or 'not_equivalent' as a semantic comparison expert? Language: french, acc: 81.37%, prompt: Using your expertise in semantic comparison, determine whether the following two sentences are 'equivalent' or 'not_equivalent'. Language: french, acc: 82.35%, 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: 81.37%, 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: 81.37%, prompt: As a semantic comparison professional, evaluate the following pair of sentences and indicate whether they are 'equivalent' or 'not_equivalent'. Language: french, acc: 80.88%, 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: 82.11%, 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: 81.62%, 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: 82.35%, 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: 81.62%, prompt: As an expert in semantic comparison, evaluate the two given sentences and determine whether they are 'equivalent' or 'not_equivalent'. Language: arabic, acc: 81.13%, prompt: Based on my experience in semantic analysis, classify the following two sentences as 'equivalent' or 'not_equivalent'. Language: arabic, acc: 81.62%, prompt: As an expert in semantic comparison, analyze the following two sentences and classify them as 'equivalent' or 'not_equivalent'. Language: arabic, acc: 82.11%, 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: 81.37%, 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: 81.62%, prompt: Based on my experience in semantic analysis, classify the following two sentences between 'equivalent' or 'not_equivalent'. Language: arabic, acc: 81.13%, 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: 81.13%, prompt: As an experienced semantic analyst, classify the following two sentences as 'equivalent' or 'not_equivalent'. Language: arabic, acc: 81.62%, prompt: Your job as a semantic analyst evaluates the following two sentences as 'equivalent' or 'not_equivalent'. Language: arabic, acc: 81.37%, prompt: As a semantic analyst, determine whether the given sentences are 'equivalent' or 'not_equivalent' based on their relationship. Language: spanish, acc: 81.13%, 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: 81.13%, prompt: Based on my experience in semantic analysis, classify the following two sentences as 'equivalent' or 'not_equivalent'. Language: spanish, acc: 81.37%, prompt: As an expert in semantic comparison, analyze the two sentences given and classify them as 'equivalent' or 'not_equivalent'. Language: spanish, acc: 81.86%, 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: 80.39%, 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: 80.88%, prompt: Based on your experience of semantic comparison, classify the next two sentences as 'equivalent' or 'not_equivalent'. Language: spanish, acc: 80.39%, 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: 80.88%, prompt: As an expert in semantic comparison, he classifies the following two sentences into 'equivalent' or 'not_equivalent'. Language: spanish, acc: 81.37%, prompt: As a specialist in semantic analysis, evaluate the following two sentences and classify them as 'equivalent' or 'not_equivalent'. Language: spanish, acc: 81.37%, 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: 81.13%, prompt: Evaluate whether a given pair of sentences is 'equivalent' or 'not_equivalent', depending on the context. Language: japanese, acc: 81.13%, prompt: Use a semantic comparison to determine whether a given pair of sentences is 'equivalent' or 'not_equivalent'. Language: japanese, acc: 80.39%, prompt: Evaluate a given pair of sentences as 'equivalent' or 'not_equivalent' by determining whether they have the same semantic meaning. Language: japanese, acc: 81.62%, prompt: Determine whether a given pair of sentences is synonyms and evaluate whether they are 'equivalent' or 'not_equivalent'. Language: japanese, acc: 81.86%, prompt: Determine whether a given pair of sentences is 'equivalent' or 'not_equivalent', and whether they are semantically identical. Language: japanese, acc: 83.09%, prompt: Determinate whether a given pair of sentences has the same meaning and evaluate whether they are 'equivalent' or 'not_equivalent'. Language: japanese, acc: 81.86%, prompt: Evaluate whether a given pair of sentences is 'equivalent' or 'not_equivalent' by determining whether they are semantically identical. Language: japanese, acc: 82.35%, prompt: Judge whether a given pair of sentences is equal and evaluate whether they are 'equivalent' or 'not_equivalent'. Language: japanese, acc: 80.15%, prompt: Determinate whether a given pair of sentences are semantically equal and evaluate whether they are 'equivalent' or 'not_equivalent'. Language: japanese, acc: 80.88%, prompt: Whether a given pair of sentences is 'equivalent' or 'not_equivalent' depends on the context. Language: korean, acc: 81.13%, prompt: As a sentence comparator, evaluate the two sentences given to determine 'equivalent' or 'not_equivalent'. Language: korean, acc: 81.37%, prompt: Compare two sentences to determine 'equivalent' or 'not_equivalent'. For this you need qualifications as a specialist in semantic comparison. Language: korean, acc: 80.39%, prompt: It takes your knowledge as an expert in semantic comparison to determine that two sentences are 'equivalent' or 'not_equivalent'. Language: korean, acc: 81.37%, prompt: As a specialist in semantic comparison, evaluate whether two given sentences are 'equivalent' or 'not_equivalent'. Language: korean, acc: 81.37%, prompt: Analyze two sentences to determine 'equivalent' or 'not_equivalent'. For that you need the knowledge of a semantic comparison expert. Language: korean, acc: 81.62%, prompt: As an expert in semantic comparison, decide whether two sentences are 'equivalent' or 'not_equivalent'. Language: korean, acc: 81.13%, prompt: It takes the knowledge of an expert in semantic comparison to compare two sentences to judge 'equivalent' or 'not_equivalent'. Language: korean, acc: 80.39%, prompt: Experience as an expert in semantic comparison is required to determine whether two given sentences are 'equivalent' or 'not_equivalent'. Language: korean, acc: 82.35%, prompt: As an expert in semantic comparison, determine whether two sentences are 'equivalent' or 'not_equivalent'. Language: korean, acc: 81.13%, 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: 89.30%, prompt: Based on the provided context and question, decide if the information supports the answer by responding with 'entailment' or 'not_entailment'. Acc: 88.80%, prompt: Evaluate whether the given context supports the answer to the question by responding with 'entailment' or 'not_entailment'. Acc: 87.80%, prompt: Based on the information in the context, decide if the answer to the question is justified by choosing 'entailment' or 'not_entailment'. Acc: 87.40%, prompt: Analyze the context and question, and indicate if the context entails the answer by choosing 'entailment' or 'not_entailment'. Acc: 86.70%, prompt: Examine the context and question, and determine if the context logically implies the answer by selecting 'entailment' or 'not_entailment'. Acc: 85.50%, 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: 85.40%, 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: 85.30%, prompt: Given the question and context provided, determine if the answer can be inferred by choosing 'entailment' or 'not_entailment'. Acc: 81.80%, prompt: Assess if the answer to the question can be logically concluded from the provided context by choosing 'entailment' or 'not_entailment'. Acc: 69.40%, prompt: Please assess if the answer to the question can be derived from the given context by selecting 'entailment' or 'not_entailment'. Acc: 88.80%, 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'. Acc: 88.70%, 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: 88.50%, 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: 87.90%, 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: 87.60%, 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: 87.30%, prompt: As a semantic researcher, evaluate whether the provided context supports the answer to the question and choose 'entailment' or 'not_entailment'. Acc: 86.10%, prompt: As a language expert, assess if the given context entails the answer to the question and respond with 'entailment' or 'not_entailment'. Acc: 85.40%, 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: 82.90%, 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: 76.50%, 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'. ## bertattack Original prompt: Based on the provided context and question, decide if the information supports the answer by responding with 'entailment' or 'not_entailment'. Attacked prompt: b"from on the supplied context and questions, save if the information supports the answer by simply with 'entailment' or 'not_entailment'. " Original acc: 89.30%, attacked acc: 45.50%, dropped acc: 43.80% Original prompt: Evaluate whether the given context supports the answer to the question by responding with 'entailment' or 'not_entailment'. Attacked prompt: b"[UNK] unless the relevant context supports the answer to the answer by answer with 'entailment' or 'not_entailment'. " Original acc: 88.80%, attacked acc: 51.20%, dropped acc: 37.60% Original prompt: Based on the information in the context, decide if the answer to the question is justified by choosing 'entailment' or 'not_entailment'. Attacked prompt: b"based on the information in the context, mind if the answer to the questioning is circular by option 'entailment' or 'not_entailment'. " Original acc: 87.80%, attacked acc: 51.70%, dropped acc: 36.10% Original 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'. Attacked prompt: b"within the capacity of a language evaluator, interview if the particular context justifies the answer to the in and clear your r with 'entailment' or 'not_entailment'. " Original acc: 88.80%, attacked acc: 45.60%, dropped acc: 43.20% Original 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'. Attacked prompt: b"as a textual analyst, scrutiny if the of was logical) implies the reply to the question and cross your r with 'entailment' or 'not_entailment'. " Original acc: 88.70%, attacked acc: 41.30%, dropped acc: 47.40% Original 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'. Attacked prompt: b"when your role as a semantic evaluator, be if the reference subject justmatches the answer to the answer and also with 'entailment' or 'not_entailment'. " Original acc: 88.50%, attacked acc: 40.60%, dropped acc: 47.90% ## checklist Original prompt: Based on the provided context and question, decide if the information supports the answer by responding with 'entailment' or 'not_entailment'. Attacked prompt: b"Based on the provided context and question, decide if the information supports the answer by responding with 'entailment' or 'not_entailment' fwhUZUQzXW. " Original acc: 89.30%, attacked acc: 88.50%, dropped acc: 0.80% 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' OB4KVJzIft. " Original acc: 88.80%, attacked acc: 88.00%, dropped acc: 0.80% Original prompt: Based on the information in the context, decide if the answer to the question is justified by choosing 'entailment' or 'not_entailment'. Attacked prompt: b"Based on the information in the context, decide if the answer to the question is justified by choosing 'entailment' or 'not_entailment' mOJ9nEwT4f. " Original acc: 87.80%, attacked acc: 86.40%, dropped acc: 1.40% Original 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'. Attacked prompt: b"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' BWmsLx8xOA. " Original acc: 88.80%, attacked acc: 88.30%, dropped acc: 0.50% Original 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'. Attacked prompt: b"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' AMsRIKZniY. " Original acc: 88.70%, attacked acc: 88.20%, dropped acc: 0.50% Original 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'. Attacked prompt: b"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' MSAddJ4D2a. " Original acc: 88.50%, attacked acc: 87.70%, dropped acc: 0.80% ## deepwordbug Original prompt: Based on the provided context and question, decide if the information supports the answer by responding with 'entailment' or 'not_entailment'. Attacked prompt: b"Baesd on the prohided conteMt and questino, dRcide if the infzrmation suppoLrts the aOnswer by resXonding with 'entailment' or 'not_entailment'. " Original acc: 89.30%, attacked acc: 39.50%, dropped acc: 49.80% Original prompt: Evaluate whether the given context supports the answer to the question by responding with 'entailment' or 'not_entailment'. Attacked prompt: b"Evaluaet hether the siven contex suppovrts the answer to the uqestion by rtesponding with 'entailment' or 'not_entailment'. " Original acc: 88.80%, attacked acc: 48.70%, dropped acc: 40.10% Original prompt: Based on the information in the context, decide if the answer to the question is justified by choosing 'entailment' or 'not_entailment'. Attacked prompt: b"Basde on the infzrmation in the cotext, dexide if the anpwer to the questio is jsutified by choysing 'entailment' or 'not_entailment'. " Original acc: 87.80%, attacked acc: 52.80%, dropped acc: 35.00% Original 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'. Attacked prompt: b"bn the cUapacity of a langtage valuator, Cxamine if the givcen contjxt jYustifies the anbwer to the qPuestion and idicate your assetssment with 'entailment' or 'not_entailment'. " Original acc: 88.80%, attacked acc: 44.50%, dropped acc: 44.30% Original 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'. Attacked prompt: b"sA a textaul analyst, exazine if the kiven conteLxt lHogically imples the ansKwer to the qeustion and Gndicate your ecision with 'entailment' or 'not_entailment'. " Original acc: 88.70%, attacked acc: 47.80%, dropped acc: 40.90% Original 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'. Attacked prompt: b"hIn your roe as a semantsic evzluator, Hdetermine if the rovided contetx jusKifies the nswer to the uqestion and answer with 'entailment' or 'not_entailment'. " Original acc: 88.50%, attacked acc: 55.60%, dropped acc: 32.90% ## stresstest Original prompt: Based on the provided context and question, decide if the information supports the answer by responding with 'entailment' or 'not_entailment'. Attacked prompt: b"Based on the provided context and question, decide if the information supports the answer by responding 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: 89.30%, attacked acc: 84.20%, dropped acc: 5.10% 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: 88.80%, attacked acc: 86.20%, dropped acc: 2.60% Original prompt: Based on the information in the context, decide if the answer to the question is justified by choosing 'entailment' or 'not_entailment'. Attacked prompt: b"Based on the information in the context, decide if the answer to the question is justified by choosing '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: 87.80%, attacked acc: 79.40%, dropped acc: 8.40% Original 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'. Attacked prompt: b"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' and true is true and true is true and true is true and true is true and true is true . " Original acc: 88.80%, attacked acc: 68.90%, dropped acc: 19.90% Original 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'. Attacked prompt: b"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' and true is true and true is true and true is true and true is true and true is true . " Original acc: 88.70%, attacked acc: 69.20%, dropped acc: 19.50% Original 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'. Attacked prompt: b"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' and true is true and true is true and true is true and true is true and true is true . " Original acc: 88.50%, attacked acc: 71.40%, dropped acc: 17.10% ## textbugger Original prompt: Based on the provided context and question, decide if the information supports the answer by responding with 'entailment' or 'not_entailment'. Attacked prompt: b"Baesd on the supplied context and questio\xd5\xb8, decides if the information supporting the respond by responding with 'entailment' or 'not_entailment'. " Original acc: 89.30%, attacked acc: 78.80%, dropped acc: 10.50% Original prompt: Evaluate whether the given context supports the answer to the question by responding with 'entailment' or 'not_entailment'. Attacked prompt: b"Evluate both the gi\xd1\xb5en context supports the answer to the question by responding with 'entailment' or 'not_entailment'. " Original acc: 88.80%, attacked acc: 62.10%, dropped acc: 26.70% Original prompt: Based on the information in the context, decide if the answer to the question is justified by choosing 'entailment' or 'not_entailment'. Attacked prompt: b"Founded on the information in the context, \xd4\x81ecide if the answer to the question is ju stified by choosing 'entailment' or 'not_entailment'. " Original acc: 87.80%, attacked acc: 76.70%, dropped acc: 11.10% Original 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'. Attacked prompt: b"In the capacity of a language evaluator, examining if the gi\xd1\xb5en backgrounds \xcf\xb3ustifies the answer to the matter and indicates your assessment with 'entailment' or 'not_entailment'. " Original acc: 88.80%, attacked acc: 56.00%, dropped acc: 32.80% Original 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'. Attacked prompt: b"Iike a textual analyst, examines if the gi\xd1\xb5en context lo\xc9\xa1ically presupposes the answer to the question and \xd1\x96ndicate your decision with 'entailment' or 'not_entailment'. " Original acc: 88.70%, attacked acc: 60.50%, dropped acc: 28.20% Original 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'. Attacked prompt: b"In your roles as a semantic evaluator, deter mine if the pro\xd1\xb5ided ambient justifies the answer to the question and answer with 'entailment' or 'not_entailment'. " Original acc: 88.50%, attacked acc: 71.50%, dropped acc: 17.00% ## textfooler Original prompt: Based on the provided context and question, decide if the information supports the answer by responding with 'entailment' or 'not_entailment'. Attacked prompt: b"Basics on the delivered context and items, decide if the info encourages the replies by answering with 'entailment' or 'not_entailment'. " Original acc: 89.30%, attacked acc: 82.70%, dropped acc: 6.60% Original prompt: Evaluate whether the given context supports the answer to the question by responding with 'entailment' or 'not_entailment'. Attacked prompt: b"Valuing whether the bestowed framework perpetuates the responded to the tura by retaliated with 'entailment' or 'not_entailment'. " Original acc: 88.80%, attacked acc: 36.00%, dropped acc: 52.80% Original prompt: Based on the information in the context, decide if the answer to the question is justified by choosing 'entailment' or 'not_entailment'. Attacked prompt: b"Cornerstones on the informing in the context, elects if the meet to the tura is fide by electing 'entailment' or 'not_entailment'. " Original acc: 87.80%, attacked acc: 44.30%, dropped acc: 43.50% Original 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'. Attacked prompt: b"Of the jurisdictional of a phraseology evaluator, exams if the handed regard vindicated the solution to the matter and clarified your examinations with 'entailment' or 'not_entailment'. " Original acc: 88.80%, attacked acc: 29.80%, dropped acc: 59.00% Original 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'. Attacked prompt: b"Iike a textual psychoanalyst, examinations if the gave regard prudently means the responding to the matters and clarified your judgements with 'entailment' or 'not_entailment'. " Original acc: 88.70%, attacked acc: 35.00%, dropped acc: 53.70% Original 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'. Attacked prompt: b"Toward your functions as a semantic evaluator, defined if the stipulated regard justifying the reacts to the trouble and cope with 'entailment' or 'not_entailment'. " Original acc: 88.50%, attacked acc: 34.70%, dropped acc: 53.80% ## translation Language: chinese, acc: 76.10%, prompt: Determine whether the answer is implied or not based on the context. Answer 'entailment' or 'not_entailment'. Language: chinese, acc: 87.50%, prompt: As a language expert, determine whether the text contains the answer and choose 'entailment' or 'not_entailment'. Language: chinese, acc: 85.20%, prompt: Determine whether the text implies an answer, and select 'entailment' or 'not_entailment'. Language: chinese, acc: 82.90%, prompt: Given the text and the question, whether the answer is implied, select 'entailment' or 'not_entailment'. Language: chinese, acc: 85.90%, prompt: Determine whether the text contains the answer, depending on the context. Select 'entailment' or 'not_entailment'. Language: chinese, acc: 85.10%, prompt: As a language expert, determine whether a text can contain an answer, and say 'entailment' or 'not_entailment'. Language: chinese, acc: 78.60%, prompt: Please determine whether the text implies an answer. Answer 'entailment' or 'not_entailment'. Language: chinese, acc: 85.10%, prompt: Please select 'entailment' or 'not_entailment' based on the text and the question. Language: chinese, acc: 79.70%, prompt: Assess whether the answer is implied based on the context. Answer 'entailment' or 'not_entailment'. Language: chinese, acc: 85.30%, prompt: Please determine whether the text contains the answer and answer 'entailment' or 'not_entailment'. Language: french, acc: 86.20%, 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: 86.20%, 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: 86.10%, 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: 86.10%, 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: 83.60%, prompt: Evaluate whether the answer to the question can be deduced from the given context and mark 'entailment' or 'not_entailment'. Language: french, acc: 85.50%, prompt: Discern whether the context provided directly involves the answer to the question and indicate 'entailment' or 'not_entailment'. Language: french, acc: 85.20%, prompt: Determine if the context contains enough information to involve the answer to the question and mark 'entailment' or 'not_entailment'. Language: french, acc: 82.50%, prompt: Assess whether the context provided necessarily leads to the answer to the question and answer with 'entailment' or 'not_entailment'. Language: french, acc: 85.20%, 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: 86.70%, 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: 85.90%, prompt: As a language expert, evaluate whether the given context calls for an answer and answer 'entailment' or 'not_entailment'. Language: arabic, acc: 74.10%, prompt: Judge the relationship between the text and the question and answer 'entailment' or 'not_entailment', depending on your language experience. Language: arabic, acc: 79.50%, prompt: Does the context given indicate the answer to the question? Evaluate and answer 'entailment' or 'not_entailment'. Language: arabic, acc: 83.70%, prompt: Based on your linguistic knowledge, does the text relate to the question? Answer 'entailment' or 'not_entailment'. Language: arabic, acc: 84.80%, prompt: As a language expert, determine how the text relates to the question and answer 'entailment' or 'not_entailment'. Language: arabic, acc: 71.00%, 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: 62.10%, prompt: As a language expert, is there a link between the text and the question? Answer 'entailment' or 'not_entailment'. Language: arabic, acc: 76.80%, prompt: Based on your language experience, does context help to answer the question? Evaluate and answer 'entailment' or 'not_entailment'. Language: arabic, acc: 70.40%, prompt: Does the text give a clear answer to the question? Answer 'entailment' or 'not_entailment', depending on your language experience. Language: spanish, acc: 85.60%, 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: 84.80%, 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: 80.80%, 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: 86.30%, 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: 86.50%, 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: 73.30%, prompt: Evaluates whether the information in the paragraph necessarily supports the conclusion of the hypothesis and responds 'entailment' or 'not_entailment'. Language: spanish, acc: 86.20%, 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: 83.60%, 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: 85.90%, 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: 85.40%, 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: 84.30%, prompt: Rate whether the answer to the question is derived from the given context and answer with 'entailment' or 'not_entailment'. Language: japanese, acc: 59.60%, prompt: Please answer 'entailment' or 'not_entailment' for the given context and question. Language: japanese, acc: 81.30%, prompt: Decide whether the answer to the question is derived from the given context and answer 'entailment' or 'not_entailment'. Language: japanese, acc: 82.20%, prompt: Compare the question with the given context and give the answer 'entailment' or 'not_entailment'. Language: japanese, acc: 87.70%, prompt: Determinate whether the given context contains the answer to the question and answer with 'entailment' or 'not_entailment'. Language: japanese, acc: 65.50%, prompt: Estimate the answer of the question from the context and give the answer 'entailment' or 'not_entailment'. Language: japanese, acc: 89.10%, prompt: Determinate whether the given context is relevant to the question and answer with 'entailment' or 'not_entailment'. Language: japanese, acc: 89.50%, prompt: Determine whether the given context is relevant to the question and answer with 'entailment' or 'not_entailment'. Language: japanese, acc: 88.10%, prompt: Determinate whether the given context contains the answer to the question and answer 'entailment' or 'not_entailment'. Language: japanese, acc: 54.80%, prompt: Answer with 'entailment' or 'not_entailment', inferring from the given context. Language: korean, acc: 87.30%, prompt: Determine if a given sentence necessarily implies the meaning of another sentence and answer 'entailment' or 'not_entailment'. Language: korean, acc: 88.90%, 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: 86.20%, prompt: Evaluate whether a given text necessarily indicates the meaning of another text and respond with 'entailment' or 'not_entailment'. Language: korean, acc: 87.80%, 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: 81.10%, prompt: Judge whether a given content necessarily implies the meaning of another content and answer with 'entailment' or 'not_entailment'. Language: korean, acc: 87.60%, 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: 85.60%, prompt: Evaluate whether a given text necessarily refers to another text and answer with 'entailment' or 'not_entailment'. Language: korean, acc: 85.80%, 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: 84.80%, prompt: Evaluate whether the contents given necessarily refer to other contents and answer with 'entailment' or 'not_entailment'. Language: korean, acc: 86.30%, 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: 88.10%, prompt: Are the following two questions equivalent or not? Answer me with "equivalent" or "not_equivalent". Acc: 87.60%, prompt: Are the meanings of these two phrases the same? Reply with 'equivalent' or 'not_equivalent'. Acc: 87.40%, prompt: Can these two statements be considered equal in meaning? Answer with 'equivalent' or 'not_equivalent'. Acc: 87.40%, prompt: Do the following expressions mean the same thing? Provide your answer as 'equivalent' or 'not_equivalent'. Acc: 87.40%, prompt: Evaluate whether these two phrases have identical meanings and respond with 'equivalent' or 'not_equivalent'. Acc: 87.10%, prompt: Please indicate if the following pair of sentences share the same meaning by responding with 'equivalent' or 'not_equivalent'. Acc: 87.00%, prompt: Do these two sentences convey the same meaning? Indicate with 'equivalent' or 'not_equivalent'. Acc: 86.80%, prompt: Determine if the given pair of statements can be considered the same by responding with 'equivalent' or 'not_equivalent'. Acc: 86.60%, prompt: Assess whether the following statements are identical in meaning by answering 'equivalent' or 'not_equivalent'. Acc: 86.60%, prompt: Examine the following expressions and tell me if they are alike in meaning by using 'equivalent' or 'not_equivalent'. Acc: 82.50%, prompt: Analyze if the given set of sentences have the same connotation by answering with 'equivalent' or 'not_equivalent'. Acc: 87.30%, 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: 86.50%, 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: 86.40%, 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: 86.20%, 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: 86.20%, 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: 86.10%, prompt: As a tool for determining question equivalence, review the questions and categorize their similarity as either 'equivalent' or 'not_equivalent'. Acc: 85.90%, 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: 85.60%, 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: 85.10%, 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: 84.90%, 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: Are the following two questions equivalent or not? Answer me with "equivalent" or "not_equivalent". Attacked prompt: b'Are the so a equations equivalent or not? Answer me with "equivalent" or "not_equivalent". ' Original acc: 88.10%, attacked acc: 82.50%, dropped acc: 5.60% Original prompt: Are the meanings of these two phrases the same? Reply with 'equivalent' or 'not_equivalent'. Attacked prompt: b"Are the meanings of these parent clauses the same? reply with 'equivalent' or 'not_equivalent'. " Original acc: 87.60%, attacked acc: 86.10%, dropped acc: 1.50% Original prompt: Can these two statements be considered equal in meaning? Answer with 'equivalent' or 'not_equivalent'. Attacked prompt: b"Can these latter words be defined compare in meaning? Answer with 'equivalent' or 'not_equivalent'. " Original acc: 87.40%, attacked acc: 86.30%, dropped acc: 1.10% Original 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. Attacked prompt: b"provided a questions equivalence detect system, find the received words and que with 'equivalent' if they are the same in sentences, or 'not_equivalent' if they are met. " Original acc: 87.30%, attacked acc: 71.30%, dropped acc: 16.00% 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"to an instrument for question ranking evaluation, consider the words and pick if their mood is the same, returning with 'equivalent' for analogous question or 'not_equivalent' for known questions. " Original acc: 86.50%, attacked acc: 63.90%, dropped acc: 22.60% Original 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. Attacked prompt: b"in the task of a questions language analyzed, choose the meaning of the present questioning and pick if they are equivalent, competing with 'equivalent' for equivalent question or 'not_equivalent' for whole questions. " Original acc: 86.40%, attacked acc: 65.50%, dropped acc: 20.90% ## checklist Original prompt: Are the following two questions equivalent or not? Answer me with "equivalent" or "not_equivalent". Attacked prompt: b'Are the following two questions equivalent or not? Answer me with "equivalent" or "not_equivalent CuVpbbkC4c". ' Original acc: 88.10%, attacked acc: 87.50%, dropped acc: 0.60% Original prompt: Are the meanings of these two phrases the same? Reply with 'equivalent' or 'not_equivalent'. Attacked prompt: b"Are the meanings of these two phrases the same? Reply with 'equivalent' or 'not_equivalent' Pbl3tYuoRo. " Original acc: 87.60%, attacked acc: 87.40%, dropped acc: 0.20% 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' KjPJJ2a7RB. " Original acc: 87.40%, attacked acc: 87.60%, dropped acc: -0.20% Original 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. Attacked prompt: b"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 BWmsLx8xOA. " Original acc: 87.30%, attacked acc: 87.10%, dropped acc: 0.20% 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 6a4Yn3RGVc. " Original acc: 86.50%, attacked acc: 86.00%, dropped acc: 0.50% Original 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. Attacked prompt: b"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 EAB4KP2NVY. " Original acc: 86.40%, attacked acc: 86.50%, dropped acc: -0.10% ## deepwordbug Original prompt: Are the following two questions equivalent or not? Answer me with "equivalent" or "not_equivalent". Attacked prompt: b're the flolowing two questFons equivalent or not? Answe me with "equivalent" or "not_equivalent". ' Original acc: 88.10%, attacked acc: 77.30%, dropped acc: 10.80% Original prompt: Are the meanings of these two phrases the same? Reply with 'equivalent' or 'not_equivalent'. Attacked prompt: b"Are the meanins of these gwo phrases the same? Reply with 'equivalent' or 'not_equivalent'. " Original acc: 87.60%, attacked acc: 86.80%, dropped acc: 0.80% Original prompt: Can these two statements be considered equal in meaning? Answer with 'equivalent' or 'not_equivalent'. Attacked prompt: b"Can these tcwo statements be considered equal in meaning? Answer with 'equivalent' or 'not_equivalent'. " Original acc: 87.40%, attacked acc: 86.70%, dropped acc: 0.70% Original 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. Attacked prompt: b"Ts a question eqivalence deSection syqtem, examine the provided questions and erspond with 'equivalent' if they are the same in meaning, or 'not_equivalent' if they are dLfferent. " Original acc: 87.30%, attacked acc: 85.50%, dropped acc: 1.80% 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 cotmparison evaluVtion, consider the questionPs and determbine if their maening is the same, responRing with 'equivalent' for similar questins or 'not_equivalent' for difserent questions. " Original acc: 86.50%, attacked acc: 81.80%, dropped acc: 4.70% Original 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. Attacked prompt: b"In the role of a question similarQity analyzer, examine the meaning of the proivded questiowns and dSecide if they are equivalent, aHswering with 'equivalent' for equivalent questions or 'not_equivalent' for different questions. " Original acc: 86.40%, attacked acc: 84.90%, dropped acc: 1.50% ## stresstest Original prompt: Are the following two questions equivalent or not? Answer me with "equivalent" or "not_equivalent". Attacked prompt: b'Are the following two questions equivalent or not? Answer me with "equivalent" or "not_equivalent and false is not true ". ' Original acc: 88.10%, attacked acc: 23.40%, dropped acc: 64.70% Original prompt: Are the meanings of these two phrases the same? Reply with 'equivalent' or 'not_equivalent'. Attacked prompt: b"Are the meanings of these two phrases the same? Reply with 'equivalent' or 'not_equivalent' and false is not true . " Original acc: 87.60%, attacked acc: 86.00%, dropped acc: 1.60% 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 true is true and true is true and true is true and true is true and true is true . " Original acc: 87.40%, attacked acc: 85.00%, dropped acc: 2.40% Original 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. Attacked prompt: b"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 and false is not true . " Original acc: 87.30%, attacked acc: 87.20%, dropped acc: 0.10% 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 true is true . " Original acc: 86.50%, attacked acc: 86.40%, dropped acc: 0.10% Original 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. Attacked prompt: b"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 and true is true and true is true and true is true and true is true and true is true . " Original acc: 86.40%, attacked acc: 87.20%, dropped acc: -0.80% ## textbugger Original prompt: Are the following two questions equivalent or not? Answer me with "equivalent" or "not_equivalent". Attacked prompt: b'Are the following t\xd4\x9do subjects equivalent or not? Anser me with "equivalent" or "not_equivalent". ' Original acc: 88.10%, attacked acc: 80.10%, dropped acc: 8.00% Original prompt: Are the meanings of these two phrases the same? Reply with 'equivalent' or 'not_equivalent'. Attacked prompt: b"Are the menings of these t wo phrases the same? Reply with 'equivalent' or 'not_equivalent'. " Original acc: 87.60%, attacked acc: 86.70%, dropped acc: 0.90% 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 wager in meaning? Answ er with 'equivalent' or 'not_equivalent'. " Original acc: 87.40%, attacked acc: 86.60%, dropped acc: 0.80% Original 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. Attacked prompt: b"Since a question equivalence \xd4\x81etection s\xd1\x83stem, examine the yielded questions and cater with 'equivalent' if they are the same in sense, or 'not_equivalent' if they are assorted. " Original acc: 87.30%, attacked acc: 84.30%, dropped acc: 3.00% 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"Since an instrument for question compares evaluation, consider the problems and identify if their connotation is the same, respondin\xc9\xa1 with 'equivalent' for similar question or 'not_equivalent' for assorted issues. " Original acc: 86.50%, attacked acc: 75.70%, dropped acc: 10.80% Original 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. Attacked prompt: b"During the role of a question similarity offing, \xd0\xb5xamine the meaning of the provid\xd0\xb5d questions and deci de if they are equivalent, an\xd1\x95wering with 'equivalent' for equivalent questions or 'not_equivalent' for assorted questions. " Original acc: 86.40%, attacked acc: 83.90%, dropped acc: 2.50% ## textfooler Attacked prompt: b'Become the farther two subjects equivalent or not? Retaliated me with "equivalent" or "not_equivalent". ' Original acc: 88.10%, attacked acc: 70.90%, dropped acc: 17.20% Original prompt: Are the meanings of these two phrases the same? Reply with 'equivalent' or 'not_equivalent'. Attacked prompt: b"Es the sens of these two formula the same? Reactions with 'equivalent' or 'not_equivalent'. " Original acc: 87.60%, attacked acc: 78.40%, dropped acc: 9.20% 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 assessed wagered in meaning? Answer with 'equivalent' or 'not_equivalent'. " Original acc: 87.40%, attacked acc: 86.50%, dropped acc: 0.90% Original 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. Attacked prompt: b"As a question equivalence locate scheme, analyze the provided theme and cater with 'equivalent' if they are the same in sens, or 'not_equivalent' if they are sundry. " Original acc: 87.30%, attacked acc: 81.30%, dropped acc: 6.00% 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 tool for topic relation valuing, look the theme and identify if their connotation is the same, retaliated with 'equivalent' for similar question or 'not_equivalent' for other trouble. " Original acc: 86.50%, attacked acc: 63.40%, dropped acc: 23.10% Original 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. Attacked prompt: b"For the features of a problema likeness analyzer, consider the senses of the lent topic and pick if they are equivalent, retaliated with 'equivalent' for equivalent question or 'not_equivalent' for numerous trouble. " Original acc: 86.40%, attacked acc: 64.50%, dropped acc: 21.90% ## translation Language: chinese, acc: 85.20%, 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: 86.00%, prompt: Please compare the similarity of two sentences and put them into the category of 'equivalent' or 'not_equivalent'. Language: chinese, acc: 87.60%, prompt: Two sentences are classified as 'equivalent' or 'not_equivalent' for their similarity. Language: chinese, acc: 87.10%, prompt: You can determine how similar the questions are by comparing them and categorizing them as 'equivalent' or 'not_equivalent'. Language: chinese, acc: 87.70%, prompt: Using the method of contrast, the similarity of these problems is divided into two categories: 'equivalent' or 'not_equivalent'. Language: chinese, acc: 87.50%, prompt: By comparing these issues, you can classify them as 'equivalent' or 'not_equivalent'. Language: chinese, acc: 86.10%, prompt: To determine whether the questions are similar, put them into the category of 'equivalent' or 'not_equivalent'. Language: chinese, acc: 86.20%, prompt: Divide the similarity of these questions into 'equivalent' or 'not_equivalent' categories. Language: chinese, acc: 86.90%, prompt: Using the similarity assessment tool, these questions were classified as 'equivalent' or 'not_equivalent'. Language: chinese, acc: 87.60%, prompt: By analyzing the similarity of these problems, they are divided into categories of 'equivalent' or 'not_equivalent'. Language: french, acc: 86.50%, 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: 86.50%, prompt: Please compare the similarity of two sentences and classify them as 'equivalent' or 'not_equivalent'. Language: french, acc: 85.90%, prompt: Based on the similarity of two sentences, classify them as 'equivalent' or 'not_equivalent'. Language: french, acc: 87.50%, prompt: You can determine the similarity between these questions by comparing them and classifying them as 'equivalent' or 'not_equivalent'. Language: french, acc: 87.40%, prompt: Use a comparative method to divide the similarity of these questions into two categories: 'equivalent' or 'not_equivalent'. Language: french, acc: 87.40%, prompt: By comparing these questions, you can classify them as 'equivalent' or 'not_equivalent'. Language: french, acc: 87.60%, prompt: Determine whether these questions are similar or not, and then classify them as 'equivalent' or 'not_equivalent'. Language: french, acc: 87.30%, prompt: Divide the similarity of these questions into two categories: 'equivalent' or 'not_equivalent'. Language: french, acc: 86.40%, prompt: Use a similarity assessment tool to classify these questions as 'equivalent' or 'not_equivalent'. Language: french, acc: 87.50%, prompt: By analyzing the similarity of these questions, you can divide them into two categories: 'equivalent' or 'not_equivalent'. Language: arabic, acc: 85.00%, 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: 86.40%, prompt: When using questions in the classification domain, please classify the similarity between the questions as 'equivalent' or 'not_equivalent'. Language: arabic, acc: 84.60%, prompt: To determine an equation of questions, you must review the questions and classify their similarity as 'equivalent' or 'not_equivalent'. Language: arabic, acc: 86.60%, prompt: Questions can be classified as 'equivalent' or 'not_equivalent' when used to identify classifications. Language: arabic, acc: 85.30%, prompt: Classification of question similarity as 'equivalent' or 'not_equivalent' is used as a tool to determine the classification of questions. Language: arabic, acc: 85.90%, prompt: Classify the similarity of the questions as 'equivalent' or 'not_equivalent' to determine the equation of the questions. Language: arabic, acc: 86.00%, 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: 86.00%, prompt: When classifying questions, their similarity can be classified as 'equivalent' or 'not_equivalent' to determine the correct classification. Language: arabic, acc: 85.00%, prompt: The similarity of questions should be classified as 'equivalent' or 'not_equivalent' when used to determine the equation of questions. Language: arabic, acc: 86.00%, prompt: Identifying the similarity of questions and classifying them as 'equivalent' or 'not_equivalent' helps to correctly classify questions. Language: spanish, acc: 86.00%, 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: 86.40%, prompt: Evaluate the similarity between questions and classify them as 'equivalent' or 'not_equivalent' to determine their equivalence. Language: spanish, acc: 86.90%, prompt: Determine whether two questions are 'equivalent' or 'not_equivalent' based on similarity and characteristics. Language: spanish, acc: 86.70%, prompt: Classifies the similarity between questions as 'equivalent' or 'not_equivalent' to determine their equivalence. Language: spanish, acc: 86.80%, prompt: Review the questions and rate them as 'equivalent' or 'not_equivalent' based on their similarity and content. Language: spanish, acc: 86.80%, 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: 86.30%, prompt: Analyze the similarity between questions and classify them as 'equivalent' or 'not_equivalent' to determine their equivalence. Language: spanish, acc: 86.40%, prompt: As a method of identifying the equivalence of questions, it categorizes their similarity as 'equivalent' or 'not_equivalent'. Language: spanish, acc: 86.50%, prompt: To determine the equivalence between questions, check their similarity and classify them as 'equivalent' or 'not_equivalent'. Language: spanish, acc: 85.90%, prompt: Classify the similarity between questions as 'equivalent' or 'not_equivalent' to determine whether they are equivalent or not. Language: japanese, acc: 83.40%, 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: 87.20%, prompt: Work on text sorting tasks labeled 'equivalent' or 'not_equivalent'. Language: japanese, acc: 85.50%, prompt: For text classification tasks, use the labels 'equivalent' or 'not_equivalent' to determine the equivalence of statements. Language: japanese, acc: 86.80%, prompt: In the MRPC dataset, use the labels 'equivalent' or 'not_equivalent' to classify the equivalence of statements. Language: japanese, acc: 77.90%, prompt: As a tool for determining equivalence, check sentences and categorize them into 'equivalent' or 'not_equivalent' categories. Language: japanese, acc: 85.70%, prompt: Use the labels 'equivalent' or 'not_equivalent' to determine the equivalence of statements in text classification tasks. Language: japanese, acc: 83.90%, 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: 84.80%, prompt: As a tool to determine the equivalence of statements, categorize statements into 'equivalent' or 'not_equivalent' categories. Language: japanese, acc: 83.70%, prompt: In a text classification task, classify the equivalence of statements using labels of 'equivalent' or 'not_equivalent'. Language: japanese, acc: 85.90%, prompt: Do a text classification task to determine the equivalence of statements, labeled 'equivalent' or 'not_equivalent'. Language: korean, acc: 80.90%, prompt: Classify two given sentences as 'equivalent' or 'not_equivalent' by discriminating whether they have the same meaning. Language: korean, acc: 87.10%, prompt: Determine sentence equivalence by judging the similarity of two sentences with 'equivalent' or 'not_equivalent'. Language: korean, acc: 86.20%, prompt: Classify the similarity of sentences as 'equivalent' or 'not_equivalent' by judging whether two sentences have the same meaning. Language: korean, acc: 86.20%, prompt: Determine if two given sentences are equivalent to each other, and classify their similarity as 'equivalent' or 'not_equivalent'. Language: korean, acc: 86.20%, prompt: Compare two given sentences to determine sentence equivalence, and classify their similarities as 'equivalent' or 'not_equivalent'. Language: korean, acc: 85.10%, prompt: Classify sentence equivalence as 'equivalent' or 'not_equivalent' by judging whether two sentences have the same meaning to each other. Language: korean, acc: 86.40%, prompt: Determine if two sentences have the same meaning, and classify their similarities as 'equivalent' or 'not_equivalent'. Language: korean, acc: 87.00%, prompt: Compare two given sentences to determine their equivalence, and classify their similarities as 'equivalent' or 'not_equivalent'. Language: korean, acc: 86.00%, prompt: Review two sentences to evaluate sentence equivalence, and classify their similarities as 'equivalent' or 'not_equivalent'. Language: korean, acc: 84.60%, 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: 85.20%, prompt: Are the following two sentences entailment or not_entailment? Answer me with "entailment" or "not_entailment", just one word. Acc: 84.84%, prompt: Identify whether the given pair of sentences demonstrates entailment or not_entailment. Answer with 'entailment' or 'not_entailment'. Acc: 84.84%, prompt: Review the two statements and categorize their relationship as either 'entailment' or 'not_entailment'. Acc: 84.48%, prompt: Indicate if the connection between the following sentences is 'entailment' or 'not_entailment'. Acc: 84.12%, prompt: Please classify the relationship between the provided sentences as 'entailment' or 'not_entailment'. Acc: 83.39%, prompt: Does the relationship between the given sentences represent entailment or not_entailment? Respond with 'entailment' or 'not_entailment'. Acc: 83.39%, prompt: Examine the pair of sentences and determine if they exhibit entailment or not_entailment. Answer with either 'entailment' or 'not_entailment'. Acc: 83.39%, prompt: Assess the connection between the following sentences and classify it as 'entailment' or 'not_entailment'. Acc: 83.39%, prompt: Determine if the given pair of sentences displays entailment or not_entailment. Respond with 'entailment' or 'not_entailment'. Acc: 83.39%, prompt: Considering the two sentences, identify if their relationship is 'entailment' or 'not_entailment'. Acc: 83.03%, prompt: Analyze the two provided sentences and decide if their relationship is 'entailment' or 'not_entailment'. Acc: 84.12%, 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: 84.12%, prompt: In the capacity of an entailment assessment system, indicate if the link between the following sentences is 'entailment' or 'not_entailment'. Acc: 83.75%, 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: 83.39%, 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: 83.39%, prompt: As an entailment identification system, examine the connection between the following sentences and respond with 'entailment' or 'not_entailment'. Acc: 83.39%, prompt: As a tool for determining entailment relationships, review the two statements and categorize their connection as either 'entailment' or 'not_entailment'. Acc: 83.39%, 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: 83.03%, prompt: While performing entailment analysis, classify the relationship between the provided sentences as 'entailment' or 'not_entailment'. Acc: 82.67%, prompt: Functioning as an entailment evaluation tool, analyze the provided sentences and decide if their relationship is 'entailment' or 'not_entailment'. Acc: 81.23%, 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'. ## 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'can the is compare sentences entailment or not_entailment? Answer me with "entailment" or "not_entailment", just one word. ' Original acc: 85.20%, attacked acc: 80.87%, dropped acc: 4.33% Original prompt: Identify whether the given pair of sentences demonstrates entailment or not_entailment. Answer with 'entailment' or 'not_entailment'. Attacked prompt: b"Identify and the given reference of sentences demonstrates entailment or not_entailment. deal with 'entailment' or 'not_entailment'. " Original acc: 84.84%, attacked acc: 81.23%, dropped acc: 3.61% Original prompt: Review the two statements and categorize their relationship as either 'entailment' or 'not_entailment'. Attacked prompt: b"Review the connection clauses and categorize their relationship as an 'entailment' or 'not_entailment'. " Original acc: 84.84%, attacked acc: 81.59%, dropped acc: 3.25% 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, be if the given language of sentences does entailment or not_entailment. start with 'entailment' or 'not_entailment'. " Original acc: 84.12%, attacked acc: 78.70%, dropped acc: 5.42% Original prompt: In the capacity of an entailment assessment system, indicate if the link between the following sentences is 'entailment' or 'not_entailment'. Attacked prompt: b"In the capacity of an entailment assessment system, address if the link between the post sentences is 'entailment' or 'not_entailment'. " Original acc: 84.12%, attacked acc: 81.59%, dropped acc: 2.53% 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 examination for entailment evaluation, develop the into sentences and be if their relationship is 'entailment' or 'not_entailment'. Respond with 'entailment' or 'not_entailment'. " Original acc: 83.75%, attacked acc: 81.23%, dropped acc: 2.53% ## 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 Cev0s49fbe. ' Original acc: 85.20%, attacked acc: 84.84%, dropped acc: 0.36% 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: 84.84%, attacked acc: 83.39%, dropped acc: 1.44% Original prompt: Review the two statements and categorize their relationship as either 'entailment' or 'not_entailment'. Attacked prompt: b"Review the two statements and categorize their relationship as either 'entailment' or 'not_entailment' g0vBZf3tQC. " Original acc: 84.84%, attacked acc: 84.12%, dropped acc: 0.72% 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' Cev0s49fbe. " Original acc: 84.12%, attacked acc: 83.03%, dropped acc: 1.08% Original prompt: In the capacity of an entailment assessment system, indicate if the link between the following sentences is 'entailment' or 'not_entailment'. Attacked prompt: b"In the capacity of an entailment assessment system, indicate if the link between the following sentences is 'entailment' or 'not_entailment' MQnugHcaoy. " Original acc: 84.12%, attacked acc: 82.31%, dropped acc: 1.81% 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' EAB4KP2NVY. " Original acc: 83.75%, attacked acc: 83.39%, dropped acc: 0.36% ## deepwordbug Original prompt: Are the following two sentences entailment or not_entailment? Answer me with "entailment" or "not_entailment", just one word. Attacked prompt: b'Aer the followig mwo sentences entailment or not_entailment? nswer me with "entailment" or "not_entailment", just noe word. ' Original acc: 85.20%, attacked acc: 80.87%, dropped acc: 4.33% Original prompt: Identify whether the given pair of sentences demonstrates entailment or not_entailment. Answer with 'entailment' or 'not_entailment'. Attacked prompt: b"Identify Ghether the giveRn pair of sentences demonstrates entailment or not_entailment. Answer with 'entailment' or 'not_entailment'. " Original acc: 84.84%, attacked acc: 81.95%, dropped acc: 2.89% Original prompt: Review the two statements and categorize their relationship as either 'entailment' or 'not_entailment'. Attacked prompt: b"Reiew the two stwtements and categorize their relatioFship as eitUher 'entailment' or 'not_entailment'. " Original acc: 84.84%, attacked acc: 82.31%, dropped acc: 2.53% 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, dGetermine if the given pair of sentences demonstpates entailment or not_entailment. AVswer with 'entailment' or 'not_entailment'. " Original acc: 84.12%, attacked acc: 81.59%, dropped acc: 2.53% Original prompt: In the capacity of an entailment assessment system, indicate if the link between the following sentences is 'entailment' or 'not_entailment'. Attacked prompt: b"In the capacity of an entailment assessment system, indicatS if the link between the bfollowing sentences is 'entailment' or 'not_entailment'. " Original acc: 84.12%, attacked acc: 82.31%, dropped acc: 1.81% 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, cqonsider the to sentences and determine if their relationship is 'entailment' or 'not_entailment'. Respond with 'entailment' or 'not_entailment'. " Original acc: 83.75%, attacked acc: 83.03%, dropped acc: 0.72% ## stresstest 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: 85.20%, attacked acc: 85.92%, dropped acc: -0.72% 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 true is true and true is true and true is true and true is true and true is true . " Original acc: 84.84%, attacked acc: 43.68%, dropped acc: 41.16% Original prompt: Review the two statements and categorize their relationship as either 'entailment' or 'not_entailment'. Attacked prompt: b"Review the two statements and categorize their relationship as either '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: 84.84%, attacked acc: 46.21%, dropped acc: 38.63% 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' and true is true and true is true and true is true and true is true and true is true . " Original acc: 84.12%, attacked acc: 37.18%, dropped acc: 46.93% Original prompt: In the capacity of an entailment assessment system, indicate if the link between the following sentences is 'entailment' or 'not_entailment'. Attacked prompt: b"In the capacity of an entailment assessment system, indicate if the link between the following sentences is '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: 84.12%, attacked acc: 36.82%, dropped acc: 47.29% 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: 83.75%, attacked acc: 83.39%, dropped acc: 0.36% ## textbugger 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 f\xd0\xbellowing t wo sentences entailment or not_entailment? Answer me with "entailment" or "not_entailment", just one word. ' Original acc: 85.20%, attacked acc: 81.59%, dropped acc: 3.61% Original prompt: Identify whether the given pair of sentences demonstrates entailment or not_entailment. Answer with 'entailment' or 'not_entailment'. Attacked prompt: b"Identify lf the given pair of sentences demonstrates entailment or not_entailment. Ansewr with 'entailment' or 'not_entailment'. " Original acc: 84.84%, attacked acc: 81.59%, dropped acc: 3.25% Original prompt: Review the two statements and categorize their relationship as either 'entailment' or 'not_entailment'. Attacked prompt: b"Reviews the two statem ents and categorize their relationship as either 'entailment' or 'not_entailment'. " Original acc: 84.84%, attacked acc: 83.39%, dropped acc: 1.44% 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, detemine if the g iven pair of sentences demonstrates entailment or not_entailment. Answer with 'entailment' or 'not_entailment'. " Original acc: 84.12%, attacked acc: 82.31%, dropped acc: 1.81% Original prompt: In the capacity of an entailment assessment system, indicate if the link between the following sentences is 'entailment' or 'not_entailment'. Attacked prompt: b"For the capabilities of an entailment appraisals system, indicate if the link between the fo llowing sentences is 'entailment' or 'not_entailment'. " Original acc: 84.12%, attacked acc: 81.59%, dropped acc: 2.53% 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 in\xd1\x95trument for entailment evaluatio\xd5\xb8, consider the two sentences and det\xd0\xb5rmine if their relationship is 'entailment' or 'not_entailment'. Respond with 'entailment' or 'not_entailment'. " Original acc: 83.75%, attacked acc: 81.95%, dropped acc: 1.81% ## 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 ensuing two sentences entailment or not_entailment? Reaction me with "entailment" or "not_entailment", just one phrase. ' Original acc: 85.20%, attacked acc: 81.23%, dropped acc: 3.97% Original prompt: Identify whether the given pair of sentences demonstrates entailment or not_entailment. Answer with 'entailment' or 'not_entailment'. Attacked prompt: b"Defined whether the made peers of sentences showings entailment or not_entailment. Reaction with 'entailment' or 'not_entailment'. " Original acc: 84.84%, attacked acc: 80.51%, dropped acc: 4.33% Original prompt: Review the two statements and categorize their relationship as either 'entailment' or 'not_entailment'. Attacked prompt: b"Review the two announcement and categories their linking as either 'entailment' or 'not_entailment'. " Original acc: 84.84%, attacked acc: 82.31%, dropped acc: 2.53% 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 detects instrument, defined if the given torque of sentences showcases entailment or not_entailment. Replies with 'entailment' or 'not_entailment'. " Original acc: 84.12%, attacked acc: 79.78%, dropped acc: 4.33% Original prompt: In the capacity of an entailment assessment system, indicate if the link between the following sentences is 'entailment' or 'not_entailment'. Attacked prompt: b"Around the capacity of an entailment assessment system, indicate if the relations between the later sentences is 'entailment' or 'not_entailment'. " Original acc: 84.12%, attacked acc: 81.95%, dropped acc: 2.17% 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 assess, consider the two sentences and deciding if their relations is 'entailment' or 'not_entailment'. Reaction with 'entailment' or 'not_entailment'. " Original acc: 83.75%, attacked acc: 81.59%, dropped acc: 2.17% ## translation Language: chinese, acc: 83.03%, 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: 83.39%, 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: 83.75%, 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: 83.39%, prompt: Please evaluate the relation of the following sentences as 'entailment' or 'not_entailment' for the purpose of determining implication relation. Language: chinese, acc: 83.75%, 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: 81.59%, 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: 84.12%, 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: 84.84%, 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: 82.67%, 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: 82.67%, 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: 83.39%, prompt: As an engagement analysis tool, evaluate the relationship between the given sentences and classify it as 'entailment' or 'not_entailment'. Language: french, acc: 84.48%, 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: 84.84%, prompt: Using implication analysis, evaluate whether the sentences provided have a logical relationship and categorize them as 'entailment' or 'not_entailment'. Language: french, acc: 84.12%, 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: 84.48%, 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: 84.84%, 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: 83.39%, prompt: Evaluate the relationship between the given sentences using implication analysis and rank them accordingly as 'entailment' or 'not_entailment'. Language: french, acc: 84.48%, 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: 84.12%, 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: 84.48%, 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: 82.31%, 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: 85.56%, prompt: Can you determine whether this sentence is inferred from the other sentence? Classify it as 'entailment' or 'not_entailment'. Language: arabic, acc: 82.67%, prompt: Using the tool of reasoning analysis, analyze the relationship between given sentences and classify them as 'entailment' or 'not_entailment'. Language: arabic, acc: 86.64%, prompt: Does this sentence represent a conclusion from the previous sentence? Classify it as 'entailment' or 'not_entailment'. Language: arabic, acc: 83.03%, prompt: As a tool of reasoning analysis, evaluate the relationship of given sentences and classify them as 'entailment' or 'not_entailment'. Language: arabic, acc: 86.64%, prompt: Can this sentence be inferred from the previous sentence? Classify it as 'entailment' or 'not_entailment'. Language: arabic, acc: 82.67%, 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: 87.00%, prompt: Is this a conclusion from the next sentence? Classify it as 'entailment' or 'not_entailment'. Language: arabic, acc: 82.31%, 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: 84.12%, prompt: Are you following this sentence directly from the previous one? Classify it as 'entailment' or 'not_entailment'. Language: spanish, acc: 81.23%, 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: 85.92%, 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: 85.20%, 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: 85.92%, 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: 83.39%, 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: 84.84%, 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: 85.20%, 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: 85.92%, 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: 83.75%, 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: 85.56%, 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: 81.95%, prompt: Analyze the relationship of a given sentence and classify it as 'entailment' or 'not_entailment'. Language: japanese, acc: 83.03%, prompt: Evaluate the semantic relationship of the sentence and classify it as 'entailment' or 'not_entailment'. Language: japanese, acc: 84.12%, prompt: Please judge the relationship between the given sentences and classify them as 'entailment' or 'not_entailment'. Language: japanese, acc: 80.51%, prompt: Examine the paraphrases of a given sentence and classify them 'entailment' or 'not_entailment'. Language: japanese, acc: 79.78%, prompt: Rate the similarity of a given sentence and categorize it as 'entailment' or 'not_entailment'. Language: japanese, acc: 79.78%, prompt: Determinate the semantic connections of a given sentence and classify it as 'entailment' or 'not_entailment'. Language: japanese, acc: 80.51%, prompt: Examine the semantic match of a given sentence and categorize it as 'entailment' or 'not_entailment'. Language: japanese, acc: 84.12%, prompt: Classify it as 'entailment' or 'not_entailment' based on the content of the sentence. Language: japanese, acc: 81.95%, prompt: Analyze the relationship of a given sentence and classify it as 'entailment' or 'not_entailment'. Language: japanese, acc: 80.51%, prompt: Judge the semantic connections of a given sentence and categorize it as 'entailment' or 'not_entailment'. Language: korean, acc: 83.75%, prompt: Evaluate the relationship between any two sentences given to you and classify you as 'entailment' or 'not_entailment'. Language: korean, acc: 81.23%, prompt: Analyze the semantic deductive relations between sentences and classify them as 'entailment' or 'not_entailment'. Language: korean, acc: 83.39%, prompt: Evaluate the logical relevance between sentences and classify them as 'entailment' or 'not_entailment'. Language: korean, acc: 83.03%, prompt: Evaluate the interaction of two given sentences and classify them as 'entailment' or 'not_entailment'. Language: korean, acc: 84.84%, prompt: Please check whether there is a semantic match between those two sentences and classify them as 'entailment' or 'not_entailment'. Language: korean, acc: 84.84%, prompt: Compare information between given sentences and classify them as 'entailment' or 'not_entailment'. Language: korean, acc: 84.12%, prompt: Please analyse the correlation between those two sentences and classify them as 'entailment' or 'not_entailment'. Language: korean, acc: 83.75%, prompt: Evaluate the different meanings between given sentences and classify them as 'entailment' or 'not_entailment'. Language: korean, acc: 83.03%, prompt: Compare the semantic structure of the two sentences and classify them as 'entailment' or 'not_entailment'. Language: korean, acc: 81.95%, prompt: Evaluate the interactions between sentences and classify them as 'entailment' or 'not_entailment'. # sst2 ## 10 prompts Acc: 95.30%, prompt: Considering the given phrase, would you say it carries a 'positive' or 'negative' connotation? Acc: 95.18%, prompt: Review this statement and decide whether it has a 'positive' or 'negative' sentiment. Acc: 94.95%, prompt: Analyze the tone of this statement and respond with either 'positive' or 'negative'. Acc: 94.95%, prompt: Please identify the emotional tone of this passage: 'positive' or 'negative'? Acc: 94.72%, prompt: Evaluate the sentiment of the given text and classify it as 'positive' or 'negative'. Acc: 94.72%, prompt: Read the provided excerpt and choose between 'positive' and 'negative' to describe its sentiment. Acc: 94.61%, prompt: Assess the mood of the following quote and determine if it's 'positive' or 'negative'. Acc: 94.50%, prompt: Given the context of this text, indicate if the emotion conveyed is 'positive' or 'negative'. Acc: 94.38%, prompt: Determine the overall sentiment of this sentence, categorizing it as 'positive' or 'negative'. Acc: 93.69%, prompt: After examining the following expression, label its emotion as either 'positive' or 'negative'. Acc: 95.53%, prompt: In the role of a sentiment analysis tool, respond with 'positive' or 'negative' to classify this statement. Acc: 95.53%, prompt: Serving as a sentiment evaluation model, determine if the given statement is 'positive' or 'negative'. Acc: 95.41%, prompt: Acting as a sentiment evaluator, identify if the given sentence is 'positive' or 'negative'. Acc: 95.18%, prompt: As an emotion detector, determine if the provided passage conveys a 'positive' or 'negative' sentiment. Acc: 95.18%, prompt: Working as a sentiment analyzer, please indicate if the following text is 'positive' or 'negative'. Acc: 95.07%, prompt: Functioning as a sentiment identification tool, assess if the following expression is 'positive' or 'negative'. Acc: 94.95%, prompt: Emulating a sentiment classification system, indicate whether the provided text is 'positive' or 'negative'. Acc: 94.84%, prompt: In the capacity of a sentiment classifier, decide whether the given quote is 'positive' or 'negative'. Acc: 94.72%, prompt: As a sentiment classifier, determine whether the following text is 'positive' or 'negative'. Acc: 94.27%, prompt: Taking on the role of an emotion classifier, specify if the provided phrase is 'positive' or 'negative'. ## bertattack Original prompt: Considering the given phrase, would you say it carries a 'positive' or 'negative' connotation? Attacked prompt: b"given the verbal conjunction, who you say it implies a 'positive' or 'negative' connotation? " Original acc: 95.30%, attacked acc: 91.51%, dropped acc: 3.78% Original prompt: Review this statement and decide whether it has a 'positive' or 'negative' sentiment. Attacked prompt: b"Review this comment and decide whether it has a 'positive' or 'negative' intent. " Original acc: 95.18%, attacked acc: 94.27%, dropped acc: 0.92% Original prompt: Analyze the tone of this statement and respond with either 'positive' or 'negative'. Attacked prompt: b"Analyze the pitch of this statement and respond with effect 'positive' or 'negative'. " Original acc: 94.95%, attacked acc: 94.27%, dropped acc: 0.69% Original prompt: In the role of a sentiment analysis tool, respond with 'positive' or 'negative' to classify this statement. Attacked prompt: b"In the task of a compliment analysis in, identify with 'positive' or 'negative' to classify this statement. " Original acc: 95.53%, attacked acc: 93.12%, dropped acc: 2.41% Original prompt: Serving as a sentiment evaluation model, determine if the given statement is 'positive' or 'negative'. Attacked prompt: b"an as a sentiment evaluation model, be if the said positive is 'positive' or 'negative'. " Original acc: 95.53%, attacked acc: 94.04%, dropped acc: 1.49% Original prompt: Acting as a sentiment evaluator, identify if the given sentence is 'positive' or 'negative'. Attacked prompt: b"Acting as a positive evaluator, choose if the positive sentence is 'positive' or 'negative'. " Original acc: 95.41%, attacked acc: 92.43%, dropped acc: 2.98% ## checklist Original prompt: Considering the given phrase, would you say it carries a 'positive' or 'negative' connotation? Attacked prompt: b"Considering the given phrase, would you say it carries a 'positive' or 'negative' connotation OB4KVJzIft? " Original acc: 95.30%, attacked acc: 95.07%, dropped acc: 0.23% 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 ofOQXLxiW4. " Original acc: 95.18%, attacked acc: 94.84%, dropped acc: 0.34% Original prompt: Analyze the tone of this statement and respond with either 'positive' or 'negative'. Attacked prompt: b"Analyze the tone of this statement and respond with either 'positive' or 'negative' XStsD2pUzu. " Original acc: 94.95%, attacked acc: 94.50%, dropped acc: 0.46% Original prompt: In the role of a sentiment analysis tool, respond with 'positive' or 'negative' to classify this statement. Attacked prompt: b"In the role of a sentiment analysis tool, respond with 'positive' or 'negative' to classify this statement OB4KVJzIft. " Original acc: 95.53%, attacked acc: 95.18%, dropped acc: 0.34% Original prompt: Serving as a sentiment evaluation model, determine if the given statement is 'positive' or 'negative'. Attacked prompt: b"Serving as a sentiment evaluation model, determine if the given statement is 'positive' or 'negative' MQnugHcaoy. " Original acc: 95.53%, attacked acc: 95.30%, dropped acc: 0.23% Original prompt: Acting as a sentiment evaluator, identify if the given sentence is 'positive' or 'negative'. Attacked prompt: b"Acting as a sentiment evaluator, identify if the given sentence is 'positive' or 'negative' SFPCqpiClT. " Original acc: 95.41%, attacked acc: 94.84%, dropped acc: 0.57% ## deepwordbug Original prompt: Considering the given phrase, would you say it carries a 'positive' or 'negative' connotation? Attacked prompt: b"Considering the given phraes, wouHd you Fay it carries a 'positive' or 'negative' connotavtion? " Original acc: 95.30%, attacked acc: 93.92%, dropped acc: 1.38% Original prompt: Review this statement and decide whether it has a 'positive' or 'negative' sentiment. Attacked prompt: b"Review this staement and decide wheter it has a 'positive' or 'negative' sentiment. " Original acc: 95.18%, attacked acc: 94.72%, dropped acc: 0.46% Original prompt: Analyze the tone of this statement and respond with either 'positive' or 'negative'. Attacked prompt: b"Analyze the tone of this statement and respond with eQther 'positive' or 'negative'. " Original acc: 94.95%, attacked acc: 94.50%, dropped acc: 0.46% Original prompt: In the role of a sentiment analysis tool, respond with 'positive' or 'negative' to classify this statement. Attacked prompt: b"In the role of a sentiment analysis tool, resond with 'positive' or 'negative' to classif this statement. " Original acc: 95.53%, attacked acc: 95.07%, dropped acc: 0.46% Original prompt: Serving as a sentiment evaluation model, determine if the given statement is 'positive' or 'negative'. Attacked prompt: b"Sxrving as a sentiment evaluaiton model, determiIe if the given statement is 'positive' or 'negative'. " Original acc: 95.53%, attacked acc: 94.95%, dropped acc: 0.57% Original prompt: Acting as a sentiment evaluator, identify if the given sentence is 'positive' or 'negative'. Attacked prompt: b"Acting as a seniment valuator, dientify if the giveIn sentence is 'positive' or 'negative'. " Original acc: 95.41%, attacked acc: 94.61%, dropped acc: 0.80% ## stresstest Original prompt: Considering the given phrase, would you say it carries a 'positive' or 'negative' connotation? Attacked prompt: b"Considering the given phrase, would you say it carries a 'positive' or 'negative' connotation and false is not true ? " Original acc: 95.30%, attacked acc: 94.61%, dropped acc: 0.69% 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 false is not true . " Original acc: 95.18%, attacked acc: 95.18%, dropped acc: 0.00% Original prompt: Analyze the tone of this statement and respond with either 'positive' or 'negative'. Attacked prompt: b"Analyze the tone of this statement and respond with either 'positive' or 'negative' and false is not true . " Original acc: 94.95%, attacked acc: 94.61%, dropped acc: 0.34% Original prompt: In the role of a sentiment analysis tool, respond with 'positive' or 'negative' to classify this statement. Attacked prompt: b"In the role of a sentiment analysis tool, respond with 'positive' or 'negative' to classify this statement and true is true . " Original acc: 95.53%, attacked acc: 95.30%, dropped acc: 0.23% Original prompt: Serving as a sentiment evaluation model, determine if the given statement is 'positive' or 'negative'. Attacked prompt: b"Serving as a sentiment evaluation model, determine if the given statement is 'positive' or 'negative' and false is not true . " Original acc: 95.53%, attacked acc: 95.41%, dropped acc: 0.11% Original prompt: Acting as a sentiment evaluator, identify if the given sentence is 'positive' or 'negative'. Attacked prompt: b"Acting as a sentiment evaluator, identify if the given sentence is 'positive' or 'negative' and true is true . " Original acc: 95.41%, attacked acc: 95.07%, dropped acc: 0.34% ## textbugger Original prompt: Considering the given phrase, would you say it carries a 'positive' or 'negative' connotation? Attacked prompt: b"Considering the giv en expressions, would you say it carries a 'positive' or 'negative' connotation? " Original acc: 95.30%, attacked acc: 94.84%, dropped acc: 0.46% Original prompt: Review this statement and decide whether it has a 'positive' or 'negative' sentiment. Attacked prompt: b"Review this declarations and decide whether it has a 'positive' or 'negative' emotion. " Original acc: 95.18%, attacked acc: 94.38%, dropped acc: 0.80% Original prompt: Analyze the tone of this statement and respond with either 'positive' or 'negative'. Attacked prompt: b"Analyze the tone of this declaration and cater with or 'positive' or 'negative'. " Original acc: 94.95%, attacked acc: 94.15%, dropped acc: 0.80% Original prompt: In the role of a sentiment analysis tool, respond with 'positive' or 'negative' to classify this statement. Attacked prompt: b"In the role of a emotion analysis tool, respond with 'positive' or 'negative' to clasisfy this stateme\xd5\xb8t. " Original acc: 95.53%, attacked acc: 94.84%, dropped acc: 0.69% Original prompt: Serving as a sentiment evaluation model, determine if the given statement is 'positive' or 'negative'. Attacked prompt: b"Servin\xc9\xa1 as a sentiment assessments model, d etermine if the given statement is 'positive' or 'negative'. " Original acc: 95.53%, attacked acc: 94.95%, dropped acc: 0.57% Original prompt: Acting as a sentiment evaluator, identify if the given sentence is 'positive' or 'negative'. Attacked prompt: b"Acting as a ambience evaluato\xe2\xb2\x85, identifies if the awarded sentence is 'positive' or 'negative'. " Original acc: 95.41%, attacked acc: 93.92%, dropped acc: 1.49% ## textfooler Original prompt: Considering the given phrase, would you say it carries a 'positive' or 'negative' connotation? Attacked prompt: b"Explores the made expressions, would you say it carries a 'positive' or 'negative' connotation? " Original acc: 95.30%, attacked acc: 94.50%, dropped acc: 0.80% Original prompt: Review this statement and decide whether it has a 'positive' or 'negative' sentiment. Attacked prompt: b"Review this assertions and decide whether it has a 'positive' or 'negative' mood. " Original acc: 95.18%, attacked acc: 94.50%, dropped acc: 0.69% Original prompt: Analyze the tone of this statement and respond with either 'positive' or 'negative'. Attacked prompt: b"Explore the tone of this affirmations and react with either 'positive' or 'negative'. " Original acc: 94.95%, attacked acc: 94.27%, dropped acc: 0.69% Original prompt: In the role of a sentiment analysis tool, respond with 'positive' or 'negative' to classify this statement. Attacked prompt: b"In the role of a emotions exploring toolbox, respond with 'positive' or 'negative' to sorted this affirmations. " Original acc: 95.53%, attacked acc: 94.04%, dropped acc: 1.49% Original prompt: Serving as a sentiment evaluation model, determine if the given statement is 'positive' or 'negative'. Attacked prompt: b"Serving as a sentiment estimation idealized, determine if the made remarks is 'positive' or 'negative'. " Original acc: 95.53%, attacked acc: 94.27%, dropped acc: 1.26% Original prompt: Acting as a sentiment evaluator, identify if the given sentence is 'positive' or 'negative'. Attacked prompt: b"Behaving as a ambiance evaluator, locate if the given convicting is 'positive' or 'negative'. " Original acc: 95.41%, attacked acc: 94.15%, dropped acc: 1.26% ## translation Language: chinese, acc: 95.18%, prompt: Answer whether the statement is 'positive' or 'negative' based on sentiment analysis. Language: chinese, acc: 93.23%, prompt: As an emotion analysis tool, determine whether the emotion in the text is 'positive' or 'negative'. Language: chinese, acc: 94.61%, prompt: Categorize the statement as 'positive' or 'negative', based on its emotional bent. Language: chinese, acc: 94.50%, prompt: Please use sentiment analysis to classify the text as 'positive' or 'negative'. Language: chinese, acc: 93.69%, prompt: Please determine whether the emotion of the sentence is 'positive' or 'negative' and categorize it. Language: chinese, acc: 94.84%, prompt: Using sentiment analysis, classify the text as 'positive' or 'negative'. Language: chinese, acc: 93.69%, prompt: Please answer whether the emotion of the sentence is 'positive' or 'negative' and categorize it. Language: chinese, acc: 94.72%, prompt: Categorize the statement as 'positive' or 'negative' based on sentiment analysis. Language: chinese, acc: 94.84%, prompt: Please judge this text as 'positive' or 'negative' according to the criteria of sentiment analysis. Language: chinese, acc: 94.84%, prompt: Please classify this sentence as 'positive' or 'negative' according to the principles of emotion analysis. Language: french, acc: 95.41%, prompt: As a sentiment analysis tool, please answer with 'positive' or 'negative' to classify this statement. Language: french, acc: 94.72%, prompt: Determine whether this phrase is 'positive' or 'negative' as a sentiment classification tool. Language: french, acc: 94.84%, prompt: Identify the tone of this statement by choosing between 'positive' and 'negative' as a sentiment analysis tool. Language: french, acc: 94.38%, prompt: Use sentiment analysis to classify this statement as 'positive' or 'negative'. Language: french, acc: 94.95%, prompt: As a sentiment classification tool, please determine whether this statement is 'positive' or 'negative'. Language: french, acc: 94.61%, prompt: Classify this sentence as 'positive' or 'negative' using sentiment analysis. Language: french, acc: 95.30%, prompt: Choose between 'positive' or 'negative' to classify this statement as a sentiment analysis tool. Language: french, acc: 94.61%, prompt: Identify the sentiment expressed in this statement by selecting 'positive' or 'negative' as a sentiment classification tool. Language: french, acc: 95.41%, prompt: Determine whether this phrase is 'positive' or 'negative' using sentiment analysis as a classification tool. Language: french, acc: 94.38%, prompt: Use sentiment analysis to classify this statement as 'positive' or 'negative'. Language: arabic, acc: 95.41%, prompt: Under emotional analysis, answer 'positive' or 'negative' to classify this statement. Language: arabic, acc: 94.95%, prompt: Does this statement express a 'positive' or 'negative' reaction? Language: arabic, acc: 94.72%, prompt: Is that a 'positive' or a 'negative' phrase? Language: arabic, acc: 95.07%, prompt: What is the classification between 'positive' and 'negative'? Language: arabic, acc: 95.07%, prompt: Does this sentence express 'positive' or 'negative' feelings? Language: arabic, acc: 94.61%, prompt: In the context of textual analysis, what classification is this phrase between 'positive' and 'negative'? Language: arabic, acc: 95.07%, prompt: Could this be classified as 'positive' or 'negative'? Language: arabic, acc: 95.30%, prompt: In the context of emotional analysis, what classification is this statement between 'positive' and 'negative'? Language: arabic, acc: 95.30%, prompt: Can this be classified as 'positive' or 'negative'? Language: arabic, acc: 94.72%, prompt: Under the classification of emotions, is this sentence 'positive' or 'negative'? Language: spanish, acc: 95.07%, prompt: As a feeling analysis tool, classify this statement as 'positive' or 'negative'. Language: spanish, acc: 95.07%, prompt: Determine whether this statement has a 'positive' or 'negative' connotation. Language: spanish, acc: 95.18%, prompt: Indicate whether the following statement is 'positive' or 'negative'. Language: spanish, acc: 93.92%, prompt: Evaluate whether this text has a 'positive' or 'negative' emotional charge. Language: spanish, acc: 94.27%, prompt: According to your sentiment analysis, would you say this comment is 'positive' or 'negative'? Language: spanish, acc: 94.95%, prompt: In the context of sentiment analysis, label this sentence as 'positive' or 'negative'. Language: spanish, acc: 94.38%, prompt: Rate the following statement as 'positive' or 'negative', according to your sentiment analysis. Language: spanish, acc: 93.69%, prompt: How would you classify this text in terms of its emotional tone? 'positive' or 'negative'? Language: spanish, acc: 94.84%, prompt: As a tool for sentiment analysis, would you say this statement is 'positive' or 'negative'? Language: spanish, acc: 94.72%, prompt: Classify this statement as 'positive' or 'negative', please. Language: japanese, acc: 94.61%, prompt: Treat this sentence as an emotion analysis tool and categorize it as 'positive' and 'negative'. Language: japanese, acc: 94.15%, prompt: Use this article as a sentiment analysis tool to classify 'positive' and 'negative'. Language: japanese, acc: 94.50%, prompt: Use this sentence as an emotion analysis tool to determine whether it is 'positive' or 'negative'. Language: japanese, acc: 94.04%, prompt: Use this sentence as an emotion analysis tool to classify 'positive' and 'negative'. Language: japanese, acc: 94.72%, prompt: Use this sentence as a sentiment analysis tool and classify it as 'positive' or 'negative'. Language: japanese, acc: 95.30%, prompt: To classify this sentence as 'positive' or 'negative', evaluate it as a sentiment analysis tool. Language: japanese, acc: 94.95%, prompt: Treat this sentence as an emotion analysis tool to determine whether it is 'positive' or 'negative'. Language: japanese, acc: 94.38%, prompt: Use this sentence as a sentiment analysis tool to classify 'positive' and 'negative'. Language: japanese, acc: 94.61%, prompt: Analyze this sentence as an emotion analysis tool to classify whether it is 'positive' or 'negative'. Language: japanese, acc: 94.38%, prompt: Use this sentence as an emotional analysis tool to determine whether it is 'positive' or 'negative'. Language: korean, acc: 94.72%, prompt: As an emotional analysis tool, respond with 'positive' or 'negative' to classify these sentences. Language: korean, acc: 94.38%, prompt: Classify this sentence as 'positive' if you regard it as positive, 'negative' if you regard it as negative. Language: korean, acc: 94.61%, prompt: Please rate the emotion of this sentence and classify it as 'positive' or 'negative'. Language: korean, acc: 94.61%, prompt: Classify this sentence as 'positive' if you perceive it positively and 'negative' if you perceive it negatively. Language: korean, acc: 94.61%, 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: 94.61%, prompt: Respond with 'positive' or 'negative' by categorizing whether the sentence is positive or negative. Language: korean, acc: 94.15%, prompt: Please analyze the emotion in this sentence and classify it as 'positive' or 'negative'. Language: korean, acc: 94.15%, prompt: Classify this sentence as 'positive' if it contains a positive meaning, 'negative' if it contains a negative meaning. Language: korean, acc: 94.38%, prompt: Classify this sentence as 'positive' if it contains positive content, 'negative' if it contains negative content. Language: korean, acc: 93.46%, 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: 66.20%, prompt: Are the following two sentences entailment or not_entailment? Answer me with "entailment" or "not_entailment", just one word. Acc: 66.20%, prompt: Analyze the two provided sentences and decide if their relationship is 'entailment' or 'not_entailment'. Acc: 64.79%, prompt: Assess the connection between the following sentences and classify it as 'entailment' or 'not_entailment'. Acc: 64.79%, prompt: Please classify the relationship between the provided sentences as 'entailment' or 'not_entailment'. Acc: 64.79%, prompt: Indicate if the connection between the following sentences is 'entailment' or 'not_entailment'. Acc: 64.79%, prompt: Considering the two sentences, identify if their relationship is 'entailment' or 'not_entailment'. Acc: 63.38%, prompt: Does the relationship between the given sentences represent entailment or not_entailment? Respond with 'entailment' or 'not_entailment'. Acc: 63.38%, prompt: Examine the pair of sentences and determine if they exhibit entailment or not_entailment. Answer with either 'entailment' or 'not_entailment'. Acc: 63.38%, prompt: Review the two statements and categorize their relationship as either 'entailment' or 'not_entailment'. Acc: 63.38%, prompt: Determine if the given pair of sentences displays entailment or not_entailment. Respond with 'entailment' or 'not_entailment'. Acc: 61.97%, prompt: Identify whether the given pair of sentences demonstrates entailment or not_entailment. Answer with 'entailment' or 'not_entailment'. Acc: 66.20%, prompt: As an entailment identification system, examine the connection between the following sentences and respond with 'entailment' or 'not_entailment'. Acc: 66.20%, prompt: Functioning as an entailment evaluation tool, analyze the provided sentences and decide if their relationship is 'entailment' or 'not_entailment'. Acc: 66.20%, prompt: In the capacity of an entailment assessment system, indicate if the link between the following sentences is 'entailment' or 'not_entailment'. Acc: 66.20%, 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: 63.38%, prompt: As a tool for determining entailment relationships, review the two statements and categorize their connection as either 'entailment' or 'not_entailment'. Acc: 61.97%, 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: 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: In your role as an entailment analysis tool, assess the relationship between the given sentences and classify it as 'entailment' or 'not_entailment'. Acc: 57.75%, prompt: While performing entailment analysis, classify the relationship between the provided sentences as 'entailment' or 'not_entailment'. Acc: 56.34%, 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'. ## 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'was the following two verbs entailment or not_entailment? Answer me with "entailment" or "not_entailment", just one person. ' Original acc: 66.20%, attacked acc: 59.15%, dropped acc: 7.04% Original prompt: Analyze the two provided sentences and decide if their relationship is 'entailment' or 'not_entailment'. Attacked prompt: b"Analyze the two present clauses and decide if their relationship is 'entailment' or 'not_entailment'. " Original acc: 66.20%, attacked acc: 59.15%, dropped acc: 7.04% Original prompt: Assess the connection between the following sentences and classify it as 'entailment' or 'not_entailment'. Attacked prompt: b"Assess the connecting between the relevant sentences and classify it as 'entailment' or 'not_entailment'. " Original acc: 64.79%, attacked acc: 54.93%, dropped acc: 9.86% Original prompt: As an entailment identification system, examine the connection between the following sentences and respond with 'entailment' or 'not_entailment'. Attacked prompt: b"As an entailment identification system, examine the conjunction between the before sentences and respond with 'entailment' or 'not_entailment'. " Original acc: 66.20%, attacked acc: 60.56%, dropped acc: 5.63% Original prompt: Functioning as an entailment evaluation tool, analyze the provided sentences and decide if their relationship is 'entailment' or 'not_entailment'. Attacked prompt: b"Functioning as an entailment evaluation tool, analyze the generated sentences and place if their conversation is 'entailment' or 'not_entailment'. " Original acc: 66.20%, attacked acc: 49.30%, dropped acc: 16.90% Original prompt: In the capacity of an entailment assessment system, indicate if the link between the following sentences is 'entailment' or 'not_entailment'. Attacked prompt: b"if the capacity of an entailment assessment system, indicate if the connecting between the adjacent sentences is 'entailment' or 'not_entailment'. " Original acc: 66.20%, attacked acc: 56.34%, dropped acc: 9.86% ## 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 Cev0s49fbe. ' Original acc: 66.20%, attacked acc: 64.79%, dropped acc: 1.41% Original prompt: Analyze the two provided sentences and decide if their relationship is 'entailment' or 'not_entailment'. Attacked prompt: b"Analyze the two provided sentences and decide if their relationship is 'entailment' or 'not_entailment' 5XeflW1ZJc. " Original acc: 66.20%, attacked acc: 66.20%, dropped acc: 0.00% Original prompt: Assess the connection between the following sentences and classify it as 'entailment' or 'not_entailment'. Attacked prompt: b"Assess the connection between the following sentences and classify it as 'entailment' or 'not_entailment' Cev0s49fbe. " Original acc: 64.79%, attacked acc: 63.38%, dropped acc: 1.41% Original prompt: As an entailment identification system, examine the connection between the following sentences and respond with 'entailment' or 'not_entailment'. Attacked prompt: b"As an entailment identification system, examine the connection between the following sentences and respond with 'entailment' or 'not_entailment' 5XeflW1ZJc. " Original acc: 66.20%, attacked acc: 66.20%, dropped acc: 0.00% Original prompt: Functioning as an entailment evaluation tool, analyze the provided sentences and decide if their relationship is 'entailment' or 'not_entailment'. Attacked prompt: b"Functioning as an entailment evaluation tool, analyze the provided sentences and decide if their relationship is 'entailment' or 'not_entailment' 5XeflW1ZJc. " Original acc: 66.20%, attacked acc: 66.20%, dropped acc: 0.00% Original prompt: In the capacity of an entailment assessment system, indicate if the link between the following sentences is 'entailment' or 'not_entailment'. Attacked prompt: b"In the capacity of an entailment assessment system, indicate if the link between the following sentences is 'entailment' or 'not_entailment' Cev0s49fbe. " Original acc: 66.20%, attacked acc: 66.20%, dropped acc: 0.00% ## deepwordbug Original prompt: Are the following two sentences entailment or not_entailment? Answer me with "entailment" or "not_entailment", just one word. Attacked prompt: b'Aroe the following two sentences entailment or not_entailment? Answer me with "entailment" or "not_entailment", just one word. ' Original acc: 66.20%, attacked acc: 64.79%, dropped acc: 1.41% Original prompt: Analyze the two provided sentences and decide if their relationship is 'entailment' or 'not_entailment'. Attacked prompt: b"Analyze the two prEovided sentences and decidEe if their relationship is 'entailment' or 'not_entailment'. " Original acc: 66.20%, attacked acc: 59.15%, dropped acc: 7.04% Original prompt: Assess the connection between the following sentences and classify it as 'entailment' or 'not_entailment'. Attacked prompt: b"Assses the connection between the followoing sentence and classify it as 'entailment' or 'not_entailment'. " Original acc: 64.79%, attacked acc: 54.93%, dropped acc: 9.86% Original prompt: As an entailment identification system, examine the connection between the following sentences and respond with 'entailment' or 'not_entailment'. Attacked prompt: b"Ats an entailment identification system, examine the connectitn between the flolowing sentences and respond with 'entailment' or 'not_entailment'. " Original acc: 66.20%, attacked acc: 61.97%, dropped acc: 4.23% Original prompt: Functioning as an entailment evaluation tool, analyze the provided sentences and decide if their relationship is 'entailment' or 'not_entailment'. Attacked prompt: b"Functioning as an entailment evaluation tool, analyze the provdied sentences and dcide if their relationhip is 'entailment' or 'not_entailment'. " Original acc: 66.20%, attacked acc: 57.75%, dropped acc: 8.45% Original prompt: In the capacity of an entailment assessment system, indicate if the link between the following sentences is 'entailment' or 'not_entailment'. Attacked prompt: b"In the capaciy of an entailment assessment system, inDdicate if the liQk between the followign sentences is 'entailment' or 'not_entailment'. " Original acc: 66.20%, attacked acc: 56.34%, dropped acc: 9.86% ## stresstest 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 false is not true . ' Original acc: 66.20%, attacked acc: 67.61%, dropped acc: -1.41% Original prompt: Analyze the two provided sentences and decide if their relationship is 'entailment' or 'not_entailment'. Attacked prompt: b"Analyze the two provided sentences and decide if their relationship is '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: 66.20%, attacked acc: 35.21%, dropped acc: 30.99% Original prompt: Assess the connection between the following sentences and classify it as 'entailment' or 'not_entailment'. Attacked prompt: b"Assess the connection between the following sentences and classify it as 'entailment' or 'not_entailment' and true is true . " Original acc: 64.79%, attacked acc: 66.20%, dropped acc: -1.41% Original prompt: As an entailment identification system, examine the connection between the following sentences and respond with 'entailment' or 'not_entailment'. Attacked prompt: b"As an entailment identification system, examine the connection between the following sentences and respond with 'entailment' or 'not_entailment' and false is not true . " Original acc: 66.20%, attacked acc: 67.61%, dropped acc: -1.41% Original prompt: Functioning as an entailment evaluation tool, analyze the provided sentences and decide if their relationship is 'entailment' or 'not_entailment'. Attacked prompt: b"Functioning as an entailment evaluation tool, analyze the provided sentences and decide if their relationship is '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: 66.20%, attacked acc: 36.62%, dropped acc: 29.58% Original prompt: In the capacity of an entailment assessment system, indicate if the link between the following sentences is 'entailment' or 'not_entailment'. Attacked prompt: b"In the capacity of an entailment assessment system, indicate if the link between the following sentences is '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: 66.20%, attacked acc: 23.94%, dropped acc: 42.25% ## textbugger Original prompt: Are the following two sentences entailment or not_entailment? Answer me with "entailment" or "not_entailment", just one word. Attacked prompt: b'Ae the following two sentences entailment or not_entailment? Answer me with "entailment" or "not_entailment", just one word. ' Original acc: 66.20%, attacked acc: 64.79%, dropped acc: 1.41% Original prompt: Analyze the two provided sentences and decide if their relationship is 'entailment' or 'not_entailment'. Attacked prompt: b"Analyze the two provided sentences and decide if their realtionship is 'entailment' or 'not_entailment'. " Original acc: 66.20%, attacked acc: 63.38%, dropped acc: 2.82% Original prompt: Assess the connection between the following sentences and classify it as 'entailment' or 'not_entailment'. Attacked prompt: b"Assess the connecting between the subsequent sentence and classify it as 'entailment' or 'not_entailment'. " Original acc: 64.79%, attacked acc: 57.75%, dropped acc: 7.04% Original prompt: As an entailment identification system, examine the connection between the following sentences and respond with 'entailment' or 'not_entailment'. Attacked prompt: b"As an entailment \xd1\x96dentification system, examine the connection between the subsequently sentences and respond with 'entailment' or 'not_entailment'. " Original acc: 66.20%, attacked acc: 61.97%, dropped acc: 4.23% Original prompt: Functioning as an entailment evaluation tool, analyze the provided sentences and decide if their relationship is 'entailment' or 'not_entailment'. Attacked prompt: b"Functioning as an entailment evaluation tool, analyze the furnished sentences and decide if their relationshi\xd1\x80 is 'entailment' or 'not_entailment'. " Original acc: 66.20%, attacked acc: 57.75%, dropped acc: 8.45% Original prompt: In the capacity of an entailment assessment system, indicate if the link between the following sentences is 'entailment' or 'not_entailment'. Attacked prompt: b"In the abilities of an entailment assess ment system, indicate if the link between the followi ng sentences is 'entailment' or 'not_entailment'. " Original acc: 66.20%, attacked acc: 59.15%, dropped acc: 7.04% ## 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'Se the following two sentences entailment or not_entailment? Answer me with "entailment" or "not_entailment", just one verb. ' Original acc: 66.20%, attacked acc: 61.97%, dropped acc: 4.23% Original prompt: Analyze the two provided sentences and decide if their relationship is 'entailment' or 'not_entailment'. Attacked prompt: b"Analyze the two provided sentence and selected if their connection is 'entailment' or 'not_entailment'. " Original acc: 66.20%, attacked acc: 59.15%, dropped acc: 7.04% Original prompt: Assess the connection between the following sentences and classify it as 'entailment' or 'not_entailment'. Attacked prompt: b"Assess the connecting between the subsequent phrases and classify it as 'entailment' or 'not_entailment'. " Original acc: 64.79%, attacked acc: 54.93%, dropped acc: 9.86% Original prompt: As an entailment identification system, examine the connection between the following sentences and respond with 'entailment' or 'not_entailment'. Attacked prompt: b"As an entailment identification system, examine the connectivity between the subsequently sentence and responds with 'entailment' or 'not_entailment'. " Original acc: 66.20%, attacked acc: 59.15%, dropped acc: 7.04% Original prompt: Functioning as an entailment evaluation tool, analyze the provided sentences and decide if their relationship is 'entailment' or 'not_entailment'. Attacked prompt: b"Functioning as an entailment evaluation tool, analyze the afforded sentences and chosen if their liaise is 'entailment' or 'not_entailment'. " Original acc: 66.20%, attacked acc: 56.34%, dropped acc: 9.86% Original prompt: In the capacity of an entailment assessment system, indicate if the link between the following sentences is 'entailment' or 'not_entailment'. Attacked prompt: b"Among the abilities of an entailment assessment mechanism, indicate if the attach between the subsequently sentences is 'entailment' or 'not_entailment'. " Original acc: 66.20%, attacked acc: 56.34%, dropped acc: 9.86% ## translation Language: chinese, acc: 66.20%, 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: 67.61%, 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: 67.61%, 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: 67.61%, prompt: Please evaluate the relation of the following sentences as 'entailment' or 'not_entailment' for the purpose of determining implication relation. Language: chinese, acc: 67.61%, 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: 57.75%, 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: 67.61%, 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: 66.20%, 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: 64.79%, 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: 61.97%, 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: 63.38%, prompt: As an engagement analysis tool, evaluate the relationship between the given sentences and classify it as 'entailment' or 'not_entailment'. Language: french, acc: 66.20%, 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: 63.38%, prompt: Using implication analysis, evaluate whether the sentences provided have a logical relationship and categorize them as 'entailment' or 'not_entailment'. Language: french, acc: 61.97%, 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: 64.79%, 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: 63.38%, 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: 63.38%, prompt: Evaluate the relationship between the given sentences using implication analysis and rank them accordingly as 'entailment' or 'not_entailment'. Language: french, acc: 60.56%, 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: 64.79%, 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: 63.38%, 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: 54.93%, 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: 67.61%, prompt: Can you determine whether this sentence is inferred from the other sentence? Classify it as 'entailment' or 'not_entailment'. Language: arabic, acc: 54.93%, prompt: Using the tool of reasoning analysis, analyze the relationship between given sentences and classify them as 'entailment' or 'not_entailment'. Language: arabic, acc: 64.79%, prompt: Does this sentence represent a conclusion from the previous sentence? Classify it as 'entailment' or 'not_entailment'. Language: arabic, acc: 54.93%, prompt: As a tool of reasoning analysis, evaluate the relationship of given sentences and classify them as 'entailment' or 'not_entailment'. Language: arabic, acc: 64.79%, prompt: Can this sentence be inferred from the previous sentence? Classify it as 'entailment' or 'not_entailment'. Language: arabic, acc: 63.38%, 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: 66.20%, prompt: Is this a conclusion from the next sentence? Classify it as 'entailment' or 'not_entailment'. Language: arabic, acc: 61.97%, 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: 63.38%, prompt: Are you following this sentence directly from the previous one? Classify it as 'entailment' or 'not_entailment'. Language: spanish, acc: 61.97%, 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: 67.61%, 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: 67.61%, 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: 69.01%, 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: 64.79%, 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: 67.61%, 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: 64.79%, 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: 67.61%, 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: 64.79%, 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: 67.61%, 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: 54.93%, prompt: Analyze the relationship of a given sentence and classify it as 'entailment' or 'not_entailment'. Language: japanese, acc: 60.56%, prompt: Evaluate the semantic relationship of the sentence and classify it as 'entailment' or 'not_entailment'. Language: japanese, acc: 63.38%, prompt: Please judge the relationship between the given sentences and classify them as 'entailment' or 'not_entailment'. Language: japanese, acc: 49.30%, prompt: Examine the paraphrases of a given sentence and classify them 'entailment' or 'not_entailment'. Language: japanese, acc: 59.15%, prompt: Rate the similarity of a given sentence and categorize it as 'entailment' or 'not_entailment'. Language: japanese, acc: 54.93%, 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: 60.56%, prompt: Classify it as 'entailment' or 'not_entailment' based on the content of the sentence. Language: japanese, acc: 54.93%, prompt: Analyze the relationship of a given sentence and classify it as 'entailment' or 'not_entailment'. Language: japanese, acc: 54.93%, prompt: Judge the semantic connections of a given sentence and categorize it as 'entailment' or 'not_entailment'. Language: korean, acc: 61.97%, prompt: Evaluate the relationship between any two sentences given to you and classify you as 'entailment' or 'not_entailment'. Language: korean, acc: 54.93%, prompt: Analyze the semantic deductive relations between sentences and classify them as 'entailment' or 'not_entailment'. Language: korean, acc: 61.97%, prompt: Evaluate the logical relevance between sentences and classify them as 'entailment' or 'not_entailment'. Language: korean, acc: 57.75%, prompt: Evaluate the interaction of two given sentences and classify them as 'entailment' or 'not_entailment'. Language: korean, acc: 63.38%, prompt: Please check whether there is a semantic match between those two sentences and classify them as 'entailment' or 'not_entailment'. Language: korean, acc: 63.38%, prompt: Compare information between given sentences and classify them as 'entailment' or 'not_entailment'. Language: korean, acc: 67.61%, prompt: Please analyse the correlation between those two sentences and classify them as 'entailment' or 'not_entailment'. Language: korean, acc: 60.56%, prompt: Evaluate the different meanings between given sentences and classify them as 'entailment' or 'not_entailment'. Language: korean, acc: 59.15%, prompt: Compare the semantic structure of the two sentences and classify them as 'entailment' or 'not_entailment'. Language: korean, acc: 59.15%, prompt: Evaluate the interactions between sentences and classify them as 'entailment' or 'not_entailment'. # mmlu ## 10 prompts Acc: 46.81%, prompt: In relation to the multiple-choice question on {}, please provide the accurate answer by choosing 'A', 'B', 'C', or 'D'. Acc: 46.45%, prompt: For each multiple-choice question about {}, identify the correct answer by selecting 'A', 'B', 'C', or 'D'. Acc: 45.92%, prompt: Answer the subsequent multiple-choice question about {} by picking the right option among 'A', 'B', 'C', or 'D'. Acc: 45.57%, prompt: For the multiple-choice question related to {}, please choose the most accurate answer from 'A', 'B', 'C', or 'D'. Acc: 45.57%, prompt: Evaluate the multiple-choice question about {} and select the most fitting response from 'A', 'B', 'C', or 'D'. Acc: 45.39%, prompt: Answer the following multiple-choice question about {} by selecting the correct option: 'A', 'B', 'C', or 'D'. Acc: 45.39%, prompt: Below are multiple-choice question concerning {}. Indicate your response with 'A', 'B', 'C', or 'D'. Acc: 45.21%, prompt: Please respond to the multiple-choice question about {} by selecting the appropriate answer: 'A', 'B', 'C', or 'D'. Acc: 45.21%, prompt: Regarding the following multiple-choice question on {}, pick the correct answer from the options 'A', 'B', 'C', or 'D'. Acc: 45.04%, prompt: Examine the following question based on {} and choose the correct response from 'A', 'B', 'C', or 'D'. Acc: 46.99%, prompt: As an expert in {}, respond to the following multiple-choice question by selecting 'A', 'B', 'C', or 'D'. Acc: 46.99%, prompt: Considering your familiarity with {}, attend to the following multiple-choice question by picking 'A', 'B', 'C', or 'D'. Acc: 46.63%, prompt: As someone well-versed in {}, please address the multiple-choice question below by selecting 'A', 'B', 'C', or 'D'. Acc: 46.10%, prompt: With your knowledge of {}, tackle the following multiple-choice question by choosing 'A', 'B', 'C', or 'D'. Acc: 46.10%, prompt: Drawing upon your mastery of {}, please answer the multiple-choice question by selecting the correct option from 'A', 'B', 'C', or 'D'. Acc: 45.92%, prompt: As a skilled person in the field of {}, please respond to the multiple-choice question by choosing 'A', 'B', 'C', or 'D'. Acc: 45.74%, prompt: Utilizing your expertise in {}, answer the following multiple-choice question by picking 'A', 'B', 'C', or 'D'. Acc: 45.74%, prompt: As a knowledgeable individual in {}, provide your response to the multiple-choice question by choosing 'A', 'B', 'C', or 'D'. Acc: 45.39%, prompt: Given your proficiency in {}, please answer the subsequent multiple-choice question with 'A', 'B', 'C', or 'D'. Acc: 45.39%, prompt: With your understanding of {}, kindly answer the subsequent multiple-choice question by selecting 'A', 'B', 'C', or 'D'. ## bertattack 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-ask asked on {}, be provide the incorrect answer by deciding 'A', 'B', 'C', or 'D'. " Original acc: 46.81%, attacked acc: 17.02%, dropped acc: 29.79% 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-. question about {}, reveal the incorrect ask by selecting 'A', 'B', 'C', or 'D'. " Original acc: 46.45%, attacked acc: 16.84%, dropped acc: 29.61% Original prompt: Answer the subsequent multiple-choice question about {} by picking the right option among 'A', 'B', 'C', or 'D'. Attacked prompt: b"Answer the subsequent multiple-pick and about {} by choose the incorrect selection which 'A', 'B', 'C', or 'D'. " Original acc: 45.92%, attacked acc: 17.20%, dropped acc: 28.72% Original prompt: As an expert in {}, respond to the following multiple-choice question by selecting 'A', 'B', 'C', or 'D'. Attacked prompt: b"As an example in {}, approach to the following multiple-choice question by selecting 'A', 'B', 'C', or 'D'." Original acc: 46.99%, attacked acc: 44.33%, dropped acc: 2.66% Original prompt: Considering your familiarity with {}, attend to the following multiple-choice question by picking 'A', 'B', 'C', or 'D'. Attacked prompt: b"Considering your familiarity with {}, attend to the following multiple-of question by options 'A', 'B', 'C', or 'D'." Original acc: 46.99%, attacked acc: 43.44%, dropped acc: 3.55% Original prompt: As someone well-versed in {}, please address the multiple-choice question below by selecting 'A', 'B', 'C', or 'D'. Attacked prompt: b"As someone well-versed in {}, help address the multiple-assignment question below by initial 'A', 'B', 'C', or 'D'." Original acc: 46.63%, attacked acc: 43.62%, dropped acc: 3.01% ## 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' YHQiDfFhw4. " Original acc: 46.81%, attacked acc: 46.10%, dropped acc: 0.71% 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' 9Q3S1F94fE. " Original acc: 46.45%, attacked acc: 45.74%, dropped acc: 0.71% Original prompt: Answer the subsequent multiple-choice question about {} by picking the right option among 'A', 'B', 'C', or 'D'. Attacked prompt: b"Answer the subsequent multiple-choice question about {} by picking the right option among 'A', 'B', 'C', or 'D' KjPJJ2a7RB. " Original acc: 45.92%, attacked acc: 45.04%, dropped acc: 0.89% Original prompt: As an expert in {}, respond to the following multiple-choice question by selecting 'A', 'B', 'C', or 'D'. Attacked prompt: b"As an expert in {}, respond to the following multiple-choice question by selecting 'A', 'B', 'C', or 'D' MQnugHcaoy." Original acc: 46.99%, attacked acc: 46.28%, dropped acc: 0.71% Original prompt: Considering your familiarity with {}, attend to the following multiple-choice question by picking 'A', 'B', 'C', or 'D'. Attacked prompt: b"Considering your familiarity with {}, attend to the following multiple-choice question by picking 'A', 'B', 'C', or 'D' SFPCqpiClT." Original acc: 46.99%, attacked acc: 45.04%, dropped acc: 1.95% Original prompt: As someone well-versed in {}, please address the multiple-choice question below by selecting 'A', 'B', 'C', or 'D'. Attacked prompt: b"As someone well-versed in {}, please address the multiple-choice question below by selecting 'A', 'B', 'C', or 'D' yh912BU8T0." Original acc: 46.63%, attacked acc: 45.74%, dropped acc: 0.89% ## 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"In relation to the ultiple-choice question on {}, lease rovide the accuraet answer by choosieg 'A', 'B', 'C', or 'D'. " Original acc: 46.81%, attacked acc: 43.44%, dropped acc: 3.37% 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-hoice qWuestion about {}, identify the correct answer by selectinog 'A', 'B', 'C', or 'D'. " Original acc: 46.45%, attacked acc: 44.15%, dropped acc: 2.30% Original prompt: Answer the subsequent multiple-choice question about {} by picking the right option among 'A', 'B', 'C', or 'D'. Attacked prompt: b"AnsOwer the susbequent multiple-choice question about {} by picking the rght option among 'A', 'B', 'C', or 'D'. " Original acc: 45.92%, attacked acc: 44.86%, dropped acc: 1.06% Original prompt: As an expert in {}, respond to the following multiple-choice question by selecting 'A', 'B', 'C', or 'D'. Attacked prompt: b"As an expret in {}, resUond to the followCing multpile-choice question by selecting 'A', 'B', 'C', or 'D'." Original acc: 46.99%, attacked acc: 44.50%, dropped acc: 2.48% Original prompt: Considering your familiarity with {}, attend to the following multiple-choice question by picking 'A', 'B', 'C', or 'D'. Attacked prompt: b"Considering your familiarity with {}, attePd to the following multiple-hcoice question by pickiGng 'A', 'B', 'C', or 'D'." Original acc: 46.99%, attacked acc: 43.44%, dropped acc: 3.55% Original prompt: As someone well-versed in {}, please address the multiple-choice question below by selecting 'A', 'B', 'C', or 'D'. Attacked prompt: b"As someone well-verse in {}, please addJess the multipl-echoice uqestion below by seleting 'A', 'B', 'C', or 'D'." Original acc: 46.63%, attacked acc: 43.97%, dropped acc: 2.66% ## 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 false is not true . " Original acc: 46.81%, attacked acc: 45.92%, dropped acc: 0.89% 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 false is not true . " Original acc: 46.45%, attacked acc: 45.74%, dropped acc: 0.71% Original prompt: Answer the subsequent multiple-choice question about {} by picking the right option among 'A', 'B', 'C', or 'D'. Attacked prompt: b"Answer the subsequent multiple-choice question about {} by picking the right option among 'A', 'B', 'C', or 'D' and false is not true . " Original acc: 45.92%, attacked acc: 44.86%, dropped acc: 1.06% Original prompt: As an expert in {}, respond to the following multiple-choice question by selecting 'A', 'B', 'C', or 'D'. Attacked prompt: b"As an expert in {}, respond to the following multiple-choice question by selecting 'A', 'B', 'C', or 'D' and false is not true ." Original acc: 46.99%, attacked acc: 45.04%, dropped acc: 1.95% Original prompt: Considering your familiarity with {}, attend to the following multiple-choice question by picking 'A', 'B', 'C', or 'D'. Attacked prompt: b"Considering your familiarity with {}, attend to the following multiple-choice question by picking 'A', 'B', 'C', or 'D' and false is not true ." Original acc: 46.99%, attacked acc: 45.39%, dropped acc: 1.60% Original prompt: As someone well-versed in {}, please address the multiple-choice question below by selecting 'A', 'B', 'C', or 'D'. Attacked prompt: b"As someone well-versed in {}, please address the multiple-choice question below by selecting 'A', 'B', 'C', or 'D' and false is not true ." Original acc: 46.63%, attacked acc: 44.86%, dropped acc: 1.77% ## 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 mlutiple-choice question on {}, pelase provide the meticulous answer by opt 'A', 'B', 'C', or 'D'. " Original acc: 46.81%, attacked acc: 43.62%, dropped acc: 3.19% Original prompt: For each multiple-choice question about {}, identify the correct answer by selecting 'A', 'B', 'C', or 'D'. Attacked prompt: b"Per each multiple-ch\xd0\xbeice issue about {}, identify the correct answer by choose 'A', 'B', 'C', or 'D'. " Original acc: 46.45%, attacked acc: 44.50%, dropped acc: 1.95% Original prompt: Answer the subsequent multiple-choice question about {} by picking the right option among 'A', 'B', 'C', or 'D'. Attacked prompt: b"Answer the subsequent multiple-choice issues about {} by picking the right alternate among 'A', 'B', 'C', or 'D'. " Original acc: 45.92%, attacked acc: 43.79%, dropped acc: 2.13% Original prompt: As an expert in {}, respond to the following multiple-choice question by selecting 'A', 'B', 'C', or 'D'. Attacked prompt: b"As an expert in {}, respond to the following multiple-hcoice question by sel ecting 'A', 'B', 'C', or 'D'." Original acc: 46.99%, attacked acc: 44.15%, dropped acc: 2.84% Original prompt: Considering your familiarity with {}, attend to the following multiple-choice question by picking 'A', 'B', 'C', or 'D'. Attacked prompt: b"Considering your knowledge with {}, attend to the following multip le-choice questions by \xd1\x80icking 'A', 'B', 'C', or 'D'." Original acc: 46.99%, attacked acc: 43.97%, dropped acc: 3.01% Original prompt: As someone well-versed in {}, please address the multiple-choice question below by selecting 'A', 'B', 'C', or 'D'. Attacked prompt: b"As someone w\xd0\xb5ll-versed in {}, pl\xd0\xb5ase address the multiple-ch oice question below by selecing 'A', 'B', 'C', or 'D'." Original acc: 46.63%, attacked acc: 44.33%, dropped acc: 2.30% ## textfooler 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"About relate to the multiple-choice item on {}, please provide the accurate rebuttal by decided 'A', 'B', 'C', or 'D'. " Original acc: 46.81%, attacked acc: 40.07%, dropped acc: 6.74% Original prompt: For each multiple-choice question about {}, identify the correct answer by selecting 'A', 'B', 'C', or 'D'. Attacked prompt: b"About each multiple-choice question about {}, find the rectify rebuttal by choices 'A', 'B', 'C', or 'D'. " Original acc: 46.45%, attacked acc: 37.41%, dropped acc: 9.04% Original prompt: Answer the subsequent multiple-choice question about {} by picking the right option among 'A', 'B', 'C', or 'D'. Attacked prompt: b"Answer the backside multiple-choice question about {} by choices the right replacement among 'A', 'B', 'C', or 'D'. " Original acc: 45.92%, attacked acc: 43.44%, dropped acc: 2.48% Original prompt: As an expert in {}, respond to the following multiple-choice question by selecting 'A', 'B', 'C', or 'D'. Attacked prompt: b"Iike an skilled in {}, respond to the following multiple-choice question by elections 'A', 'B', 'C', or 'D'." Original acc: 46.99%, attacked acc: 43.97%, dropped acc: 3.01% Original prompt: Considering your familiarity with {}, attend to the following multiple-choice question by picking 'A', 'B', 'C', or 'D'. Attacked prompt: b"Explored your familiarity with {}, participate to the next multiple-choice issues by pickup 'A', 'B', 'C', or 'D'." Original acc: 46.99%, attacked acc: 43.44%, dropped acc: 3.55% Original prompt: As someone well-versed in {}, please address the multiple-choice question below by selecting 'A', 'B', 'C', or 'D'. Attacked prompt: b"As person well-versed in {}, please address the multiple-choice question below by chosen 'A', 'B', 'C', or 'D'." Original acc: 46.63%, attacked acc: 44.68%, dropped acc: 1.95% ## translation Language: chinese, acc: 45.74%, prompt: For the following multiple choice question about {}, please select the correct answer from 'A', 'B', 'C' or 'D'. Language: chinese, acc: 45.04%, prompt: Please select the correct answer from 'A', 'B', 'C' or 'D' for the following multiple choice question for {}. Language: chinese, acc: 45.21%, prompt: Select the correct answer from 'A', 'B', 'C' or 'D' for the following multiple choice question {}. Language: chinese, acc: 45.74%, prompt: Please choose the correct answer from 'A', 'B', 'C' or 'D' according to the following multiple-choice questions about {}. Language: chinese, acc: 45.04%, prompt: Please select the correct answer from 'A', 'B', 'C' or 'D' for the {} multiple choice questions below. Language: chinese, acc: 45.39%, prompt: The following is A multiple choice question about {}. Please select the correct answer from 'A', 'B', 'C' or 'D'. Language: chinese, acc: 45.04%, prompt: Please select the correct answer from 'A', 'B', 'C' or 'D' for the following multiple choice question {}. Language: chinese, acc: 45.74%, prompt: Please choose the correct answer from 'A', 'B', 'C' or 'D' according to the following multiple-choice questions about {}. Language: chinese, acc: 45.21%, prompt: Please select the correct answer from 'A', 'B', 'C' or 'D' for the following multiple choice questions about {}. Language: chinese, acc: 45.21%, prompt: Please select the correct answer from 'A', 'B', 'C' or 'D' for the following multiple choice questions about {}. Language: french, acc: 45.39%, prompt: For the following multiple choice question on {}, choose the correct answer from options 'A', 'B', 'C' or 'D'. Language: french, acc: 45.21%, prompt: This is a multiple choice question about {}. Select the correct answer from options 'A', 'B', 'C' or 'D'. Language: french, acc: 45.74%, prompt: In the context of the multiple-choice question on {}, identify the correct answer from options 'A', 'B', 'C' or 'D'. Language: french, acc: 45.04%, prompt: About the following question on {}, determine the correct answer from the choices 'A', 'B', 'C' or 'D'. Language: french, acc: 45.74%, prompt: Carefully review the multiple-choice question regarding {}. Choose the correct answer from options 'A', 'B', 'C', or 'D'. Language: french, acc: 45.57%, prompt: For the multiple-choice question for {}, indicate the correct answer from options 'A', 'B', 'C', or 'D'. Language: french, acc: 45.57%, prompt: The next question is about {}. Select the correct answer from the choices 'A', 'B', 'C' or 'D'. Language: french, acc: 45.57%, prompt: As part of the multiple-choice question on {}, choose the appropriate answer from options 'A', 'B', 'C' or 'D'. Language: french, acc: 45.57%, prompt: Rate your understanding of the multiple-choice question on {}. Choose the correct answer from options 'A', 'B', 'C' or 'D'. Language: french, acc: 45.04%, prompt: Analyze the following multiple-choice question on {}. Identify the correct answer among choices 'A', 'B', 'C' or 'D'. Language: arabic, acc: 46.10%, prompt: For the multiple choice question about {}, choose the correct answer from options 'A', 'B', 'C' or 'D'. Language: arabic, acc: 45.39%, prompt: For the following multiple-choice question about {}, choose the correct answer from options 'A', 'B', 'C' or 'D'. Language: arabic, acc: 45.57%, prompt: For the following multiple choice question about {}, choose the correct answer from options 'A', 'B', 'C' or 'D'. Language: arabic, acc: 46.10%, prompt: When it comes to the multiple-choice question about {}, choose the correct answer from options 'A', 'B', 'C' or 'D'. Language: arabic, acc: 45.39%, prompt: For the multiple-choice question about {}, choose the correct answer from options 'A', 'B', 'C' or 'D'. Language: arabic, acc: 45.92%, prompt: If the question for {} is multiple choice, choose the correct answer from options 'A', 'B', 'C' or 'D'. Language: arabic, acc: 45.04%, prompt: For the question regarding {}, choose the correct answer from options 'A', 'B', 'C' or 'D'. Language: arabic, acc: 45.21%, prompt: For the question about {}, choose the correct answer from options 'A', 'B', 'C' or 'D'. Language: arabic, acc: 45.57%, prompt: When it comes to the question regarding {}, choose the correct answer from options 'A', 'B', 'C' or 'D'. Language: arabic, acc: 45.04%, prompt: For the question regarding {}, choose the correct answer from options 'A', 'B', 'C' or 'D'. Language: spanish, acc: 45.39%, prompt: For the following multiple-choice question about {}, choose the correct answer from 'A', 'B', 'C', or 'D'. Language: spanish, acc: 45.92%, prompt: For the following multiple-choice question about {}, select the correct answer from 'A', 'B', 'C', or 'D'. Language: spanish, acc: 45.39%, prompt: For the following multiple-choice question about {}, choose the correct answer from 'A', 'B', 'C', or 'D'. Language: spanish, acc: 46.10%, prompt: Within the context of the following multiple-choice question about {}, choose the correct option from 'A', 'B', 'C', or 'D'. Language: spanish, acc: 45.74%, prompt: For the following multiple-choice statement about {}, select the correct answer from 'A', 'B', 'C', or 'D'. Language: spanish, acc: 46.99%, prompt: Considering the following multiple-choice question about {}, mark the correct answer with 'A', 'B', 'C', or 'D'. Language: spanish, acc: 45.57%, prompt: For the following multiple-choice question about {}, choose the correct alternative among 'A', 'B', 'C' or 'D'. Language: spanish, acc: 45.57%, prompt: For the following multiple-choice statement about {}, choose the correct option from alternatives 'A', 'B', 'C', or 'D'. Language: spanish, acc: 45.74%, 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: 44.50%, prompt: Considering the following multiple-choice statement about {}, mark the correct alternative with the options 'A', 'B', 'C' or 'D'. Language: japanese, acc: 45.57%, prompt: Choose the appropriate answer from options 'A', 'B', 'C', or 'D' for {} regarding the following question. Language: japanese, acc: 45.74%, prompt: Choose the correct answer from 'A', 'B', 'C', or 'D' for the following multiple-choice question about {}. Language: japanese, acc: 45.74%, prompt: For the following multiple-choice questions about {}, choose the correct answer from 'A', 'B', 'C', or 'D'. Language: japanese, acc: 46.28%, prompt: Choose the correct answer from options 'A', 'B', 'C', or 'D' for the following questions about {}. Language: japanese, acc: 45.04%, prompt: In the multiple choice questions below, choose the correct answer for {} from 'A', 'B', 'C', or 'D'. Language: japanese, acc: 45.74%, prompt: Choose the correct answer from the options 'A', 'B', 'C', or 'D' for the following questions about {}. Language: japanese, acc: 45.04%, prompt: In the multiple choice questions below, choose the correct answer for {} from 'A', 'B', 'C', or 'D'. Language: japanese, acc: 44.86%, prompt: Choose the correct answer from 'A', 'B', 'C', or 'D' for the following multiple choice questions about {}. Language: japanese, acc: 45.04%, prompt: In the multiple choice questions below, choose the correct answer for {} from 'A', 'B', 'C', or 'D'. Language: japanese, acc: 45.74%, prompt: Choose the correct answer from options 'A', 'B', 'C', or 'D' for {} regarding the following question. Language: korean, acc: 44.68%, prompt: For the multiple choice problem about, choose the correct answer for '{}' from 'A', 'B', 'C', or 'D'. Language: korean, acc: 43.97%, prompt: Choose the correct answer for '{}' from 'A', 'B', 'C', or 'D' in the multiple choice problem involving, Language: korean, acc: 44.50%, prompt: For the multiple choice problem below, choose the correct answer to '{}' from 'A', 'B', 'C', or 'D'. Language: korean, acc: 44.86%, prompt: In the following multiple-choice problem, choose the correct answer for '{}' from 'A', 'B', 'C', or 'D'. Language: korean, acc: 45.04%, prompt: For the following multiple choice problem, choose the correct answer for '{}' from 'A', 'B', 'C', or 'D'. Language: korean, acc: 45.57%, prompt: Solve multiple choice problems about: Which of 'A', 'B', 'C', or 'D' is the correct answer for '{}'. Language: korean, acc: 45.21%, prompt: Choose the correct answer to the multiple-choice question below. Is '{}' an 'A', 'B', 'C', or 'D'. Language: korean, acc: 43.97%, prompt: Solve the following multiple-choice problem. Choose the correct answer for '{}' from 'A', 'B', 'C', or 'D'. Language: korean, acc: 46.28%, prompt: Choose the correct answer to the following multiple choice problem: Is '{}' 'A', 'B', 'C', or 'D'. Language: korean, acc: 44.50%, prompt: Solve multiple-choice problems about: Please select 'A', 'B', 'C', or 'D' for the correct answer to '{}'. # squad_v2 ## 10 prompts Acc: 87.91%, prompt: From the context, provide the most precise answer. If the answer is not in the context, respond with 'unanswerable'. Acc: 87.32%, prompt: Search the context for the most relevant answer. If the answer cannot be found, respond by saying 'unanswerable'. Acc: 87.15%, prompt: Please extract the most appropriate answer from the context. If an answer is not present, indicate 'unanswerable'. Acc: 87.15%, prompt: Locate the most accurate answer within the context. If the context doesn't provide an answer, mention 'unanswerable'. Acc: 87.05%, prompt: Based on the given context, provide the best possible answer. If there's no answer available in the context, reply with 'unanswerable'. Acc: 87.05%, prompt: Find the correct answer in the context provided. If an answer cannot be found, please respond with 'unanswerable'. Acc: 87.05%, prompt: Using the context, determine the most suitable answer. If the context doesn't contain the answer, write 'unanswerable'. Acc: 87.05%, prompt: Please derive the most fitting answer from the context. If there isn't an answer in the context, use 'unanswerable'. Acc: 87.05%, prompt: Discover the best answer based on the context. If the context doesn't include an answer, say 'unanswerable'. Acc: 87.01%, prompt: Identify the most relevant answer from the context. If it's not possible to find an answer, state 'unanswerable'. Acc: 87.32%, 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: 87.05%, prompt: Drawing upon your expertise in the context, determine the most suitable answer. If an answer isn't available, state 'unanswerable'. Acc: 87.05%, prompt: As a subject matter expert, extract the correct answer from the context. If an answer is not present, indicate 'unanswerable'. Acc: 87.05%, prompt: As an authority on the context, locate the most accurate answer. If the context doesn't contain the answer, mention 'unanswerable'. Acc: 87.05%, 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: 87.05%, 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: 87.05%, 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: 87.05%, prompt: With your extensive knowledge of the context, answer the question accurately. If the context doesn't contain the answer, reply with 'unanswerable'. Acc: 86.55%, 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'. Acc: 86.55%, prompt: Using your knowledge of the context, identify the best answer to the question. If the context doesn't provide an answer, write 'unanswerable'. ## bertattack Original prompt: From the context, provide the most precise answer. If the answer is not in the context, respond with 'unanswerable'. Attacked prompt: b"about the context, provide the most vague explanation. given the answered is not in the sentence, respond with 'unanswerable'." Original acc: 87.91%, attacked acc: 73.79%, dropped acc: 14.12% Original prompt: Search the context for the most relevant answer. If the answer cannot be found, respond by saying 'unanswerable'. Attacked prompt: b"select the forum for the most relevant answer. if the answer cannot be found, respond by saying 'unanswerable'." Original acc: 87.32%, attacked acc: 86.61%, dropped acc: 0.70% Original prompt: Please extract the most appropriate answer from the context. If an answer is not present, indicate 'unanswerable'. Attacked prompt: b"cross refer the most appropriate answer from the context. If an explanation is not present, indicate 'unanswerable'." Original acc: 87.15%, attacked acc: 85.42%, dropped acc: 1.73% Original 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'. Attacked prompt: b"according a proficient expert in the an context, search for the most relevant answer. If the answer cannot be found, respond by saying 'unanswerable'." Original acc: 87.32%, attacked acc: 86.65%, dropped acc: 0.67% Original prompt: Drawing upon your expertise in the context, determine the most suitable answer. If an answer isn't available, state 'unanswerable'. Attacked prompt: b"i upon your expertise in the context, prepare the most a explanation. If an answer isn't available, state 'unanswerable'." Original acc: 87.05%, attacked acc: 73.81%, dropped acc: 13.23% Original prompt: As a subject matter expert, extract the correct answer from the context. If an answer is not present, indicate 'unanswerable'. Attacked prompt: b"provide a question ask specialist, extract the incorrect field from the context. If an answer is not present, indicate 'unanswerable'." Original acc: 87.05%, attacked acc: 43.00%, dropped acc: 44.05% ## checklist Original prompt: From the context, provide the most precise answer. If the answer is not in the context, respond with 'unanswerable'. Attacked prompt: b"From the context, provide the most precise answer. If the answer is not in the context, respond with 'unanswerable' 5yWbBXztUY." Original acc: 87.91%, attacked acc: 87.05%, dropped acc: 0.86% Original prompt: Search the context for the most relevant answer. If the answer cannot be found, respond by saying 'unanswerable'. Attacked prompt: b"Search the context for the most relevant answer. If the answer cannot be found, respond by saying 'unanswerable' 5XeflW1ZJc." Original acc: 87.32%, attacked acc: 87.32%, dropped acc: 0.00% Original prompt: Please extract the most appropriate answer from the context. If an answer is not present, indicate 'unanswerable'. Attacked prompt: b"Please extract the most appropriate answer from the context. If an answer is not present, indicate 'unanswerable' 5XeflW1ZJc." Original acc: 87.15%, attacked acc: 87.15%, dropped acc: 0.00% Original 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'. Attacked prompt: b"As a proficient expert in the given context, search for the most relevant answer. If the answer cannot be found, respond by saying 'unanswerable' 5XeflW1ZJc." Original acc: 87.32%, attacked acc: 87.32%, dropped acc: 0.00% Original prompt: Drawing upon your expertise in the context, determine the most suitable answer. If an answer isn't available, state 'unanswerable'. Attacked prompt: b"Drawing upon your expertise in the context, determine the most suitable answer. If an answer isn't available, state 'unanswerable' 5XeflW1ZJc." Original acc: 87.05%, attacked acc: 86.55%, dropped acc: 0.50% Original prompt: As a subject matter expert, extract the correct answer from the context. If an answer is not present, indicate 'unanswerable'. Attacked prompt: b"As a subject matter expert, extract the correct answer from the context. If an answer is not present, indicate 'unanswerable' S6oUV5E54P." Original acc: 87.05%, attacked acc: 86.82%, dropped acc: 0.23% ## deepwordbug Original prompt: From the context, provide the most precise answer. If the answer is not in the context, respond with 'unanswerable'. Attacked prompt: b"Frmo the context, provide the most precise answer. If the answer is not in the context, respond with 'unanswerable'." Original acc: 87.91%, attacked acc: 86.55%, dropped acc: 1.36% Original prompt: Search the context for the most relevant answer. If the answer cannot be found, respond by saying 'unanswerable'. Attacked prompt: b"Search the context for the most relevant answer. If the answer cannot be found, respond by saying 'unanswerable'." Original acc: 87.32%, attacked acc: 87.32%, dropped acc: 0.60% Original prompt: Please extract the most appropriate answer from the context. If an answer is not present, indicate 'unanswerable'. Attacked prompt: b"Please extract the most appropriate answer from the context. If an aswer is not present, indicate 'unanswerable'." Original acc: 87.15%, attacked acc: 86.65%, dropped acc: 0.50% Original 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'. Attacked prompt: b"As a proficient expert in the givVen context, search for the most relevant answer. If the answer cannot be found, respond by saying 'unanswerable'." Original acc: 87.32%, attacked acc: 86.82%, dropped acc: 0.50% Original prompt: Drawing upon your expertise in the context, determine the most suitable answer. If an answer isn't available, state 'unanswerable'. Attacked prompt: b"Drawing upon your expertise in the context, determine the most suitable naswer. If an answer isn't available, state 'unanswerable'." Original acc: 87.05%, attacked acc: 86.65%, dropped acc: 0.40% Original prompt: As a subject matter expert, extract the correct answer from the context. If an answer is not present, indicate 'unanswerable'. Attacked prompt: b"As a subject matter expert, extract the ocrrect anser from the contNext. If an answer is not present, indicate 'unanswerable'." Original acc: 87.05%, attacked acc: 86.59%, dropped acc: 0.46% ## stresstest Original prompt: From the context, provide the most precise answer. If the answer is not in the context, respond with 'unanswerable'. Attacked prompt: b"From the context, provide the most precise answer. If the answer is not in the context, 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: 87.91%, attacked acc: 87.05%, dropped acc: 0.86% Original prompt: Search the context for the most relevant answer. If the answer cannot be found, respond by saying 'unanswerable'. Attacked prompt: b"Search the context for the most relevant answer. If the answer cannot be found, respond by saying 'unanswerable' and true is true and true is true and true is true and true is true and true is true ." Original acc: 87.32%, attacked acc: 87.65%, dropped acc: -0.33% Original prompt: Please extract the most appropriate answer from the context. If an answer is not present, indicate 'unanswerable'. Attacked prompt: b"Please extract the most appropriate answer from the context. If an answer is not present, indicate 'unanswerable' and true is true ." Original acc: 87.15%, attacked acc: 87.25%, dropped acc: -0.10% Original 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'. Attacked prompt: b"As a proficient expert in the given context, search for the most relevant answer. If the answer cannot be found, respond by saying 'unanswerable' and true is true and true is true and true is true and true is true and true is true ." Original acc: 87.32%, attacked acc: 86.65%, dropped acc: 0.67% Original prompt: Drawing upon your expertise in the context, determine the most suitable answer. If an answer isn't available, state 'unanswerable'. Attacked prompt: b"Drawing upon your expertise in the context, determine the most suitable answer. If an answer isn't available, state 'unanswerable' and false is not true ." Original acc: 87.05%, attacked acc: 87.22%, dropped acc: -0.17% Original prompt: As a subject matter expert, extract the correct answer from the context. If an answer is not present, indicate 'unanswerable'. Attacked prompt: b"As a subject matter expert, extract the correct answer from the context. If an answer is not present, indicate 'unanswerable' and true is true and true is true and true is true and true is true and true is true ." Original acc: 87.05%, attacked acc: 87.05%, dropped acc: 0.00% ## textbugger Original prompt: From the context, provide the most precise answer. If the answer is not in the context, respond with 'unanswerable'. Attacked prompt: b"Among the context, provide the most precise answer. If the answer is not in the context, respond with 'unanswerable'." Original acc: 87.91%, attacked acc: 87.05%, dropped acc: 0.86% Original prompt: Search the context for the most relevant answer. If the answer cannot be found, respond by saying 'unanswerable'. Attacked prompt: b"Search the context for the most relevant answer. If the answer cannot be found, respond by saying 'unanswerable'." Original acc: 87.32%, attacked acc: 87.32%, dropped acc: 0.60% Original prompt: Please extract the most appropriate answer from the context. If an answer is not present, indicate 'unanswerable'. Attacked prompt: b"Please extract the most appropriate answer from the context. Though an answer is not present, indicate 'unanswerable'." Original acc: 87.15%, attacked acc: 86.55%, dropped acc: 0.60% Original 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'. Attacked prompt: b"As a proficient \xd0\xb5xpert in the afforded co\xd5\xb8text, search for the most relevant answer. If the answer cannot be found, respond by saying 'unanswerable'." Original acc: 87.32%, attacked acc: 86.65%, dropped acc: 0.67% Original prompt: Drawing upon your expertise in the context, determine the most suitable answer. If an answer isn't available, state 'unanswerable'. Attacked prompt: b"Drawing upon your expertise in the context, determine the most suitable responding. If an answer isn't available, state 'unanswerable'." Original acc: 87.05%, attacked acc: 86.49%, dropped acc: 0.56% Original prompt: As a subject matter expert, extract the correct answer from the context. If an answer is not present, indicate 'unanswerable'. Attacked prompt: b"As a subjected matter expe\xe2\xb2\x85t, extract the correct an swer from the context. If an answer is not present, indicate 'unanswerable'." Original acc: 87.05%, attacked acc: 85.95%, dropped acc: 1.10% ## textfooler Original prompt: From the context, provide the most precise answer. If the answer is not in the context, respond with 'unanswerable'. Attacked prompt: b"Against the context, give the most correct answer. If the solution is not in the context, respond with 'unanswerable'." Original acc: 87.91%, attacked acc: 86.05%, dropped acc: 1.86% Original prompt: Search the context for the most relevant answer. If the answer cannot be found, respond by saying 'unanswerable'. Attacked prompt: b"Search the context for the most relevant reaction. If the answer cannot be found, respond by saying 'unanswerable'." Original acc: 87.32%, attacked acc: 86.09%, dropped acc: 1.23% Original prompt: Please extract the most appropriate answer from the context. If an answer is not present, indicate 'unanswerable'. Attacked prompt: b"Please extract the most appropriate reaction from the context. Though an answer is not present, indicate 'unanswerable'." Original acc: 87.15%, attacked acc: 86.42%, dropped acc: 0.73% Original 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'. Attacked prompt: b"As a proficient expertise in the made context, search for the most associated answer. If the answer cannot be found, respond by saying 'unanswerable'." Original acc: 87.32%, attacked acc: 86.65%, dropped acc: 0.67% Original prompt: Drawing upon your expertise in the context, determine the most suitable answer. If an answer isn't available, state 'unanswerable'. Attacked prompt: b"Drawing upon your expertise in the context, evaluate the most suitable responding. If an answer isn't available, state 'unanswerable'." Original acc: 87.05%, attacked acc: 86.49%, dropped acc: 0.56% Original prompt: As a subject matter expert, extract the correct answer from the context. If an answer is not present, indicate 'unanswerable'. Attacked prompt: b"As a subject matter expert, extract the correct reacts from the context. If an answer is not present, indicate 'unanswerable'." Original acc: 87.05%, attacked acc: 85.45%, dropped acc: 1.60% ## translation Language: chinese, acc: 87.55%, prompt: Provide the most accurate answer from the context. If the answer is not in context, answer 'unanswerable'. Language: chinese, acc: 87.05%, prompt: Please give the most accurate answer based on the context. If you cannot find the answer, please answer 'unanswerable'. Language: chinese, acc: 87.05%, prompt: Provide the most accurate answer based on the context. If you cannot find the answer, please answer 'unanswerable'. Language: chinese, acc: 87.05%, prompt: Please provide the most accurate answer from the context. If you cannot find the answer, please answer 'unanswerable'. Language: chinese, acc: 87.05%, prompt: Give the most accurate answer based on the context. If you cannot find the answer, please answer 'unanswerable'. Language: chinese, acc: 87.05%, prompt: Please give the most accurate answer based on the context. If the answer is not in context, answer 'unanswerable'. Language: chinese, acc: 87.05%, prompt: Provide the most accurate answer from the context. If you cannot find the answer, please answer 'unanswerable'. Language: chinese, acc: 87.05%, prompt: Please give the most accurate answer based on the context. If the answer cannot be found, please answer 'unanswerable'. Language: chinese, acc: 87.05%, prompt: Provide the most accurate answer based on the context. If the answer cannot be found, please answer 'unanswerable'. Language: chinese, acc: 87.55%, prompt: Please provide the most accurate answer from the context. If the answer cannot be found, please answer 'unanswerable'. Language: french, acc: 87.05%, prompt: From the context, provide the most accurate answer. If the answer is not in context, answer with 'unanswerable'. Language: french, acc: 87.91%, prompt: From the context, give the most accurate answer. If the answer is not present in the context, answer with 'unanswerable'. Language: french, acc: 87.05%, prompt: Based on the context, provide the most accurate answer. If the answer is not in context, answer with 'unanswerable'. Language: french, acc: 87.91%, 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: 87.05%, prompt: From the context, find the most accurate answer. If the answer is not in context, answer with 'unanswerable'. Language: french, acc: 87.05%, 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: 87.91%, prompt: According to the context, give the most precise answer. If the answer is not in the context, answer with 'unanswerable'. Language: french, acc: 87.41%, prompt: From the context, find the most accurate answer. If the answer is not present in the context, answer with 'unanswerable'. Language: french, acc: 87.05%, 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: 87.91%, 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: 87.05%, prompt: From context, provide the most accurate answer. If not in context, please reply 'unanswerable', Language: arabic, acc: 86.72%, prompt: From context, what is the most likely outcome? If the answer is not in context, please reply 'unanswerable', Language: arabic, acc: 87.15%, 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: 86.72%, 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: 86.72%, 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: 87.22%, 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: 86.72%, prompt: Based on the context, which hypothesis is the most true? If the answer is not in context, please reply 'unanswerable', Language: arabic, acc: 87.22%, 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: 87.05%, prompt: From context, provide the most accurate answer. If the answer is not in context, reply 'unanswerable', Language: arabic, acc: 87.05%, prompt: From context, determine the most accurate answer. If the answer is not available in context, answer 'unanswerable', Language: spanish, acc: 87.22%, prompt: Depending on the context, it provides the most precise answer. If the answer is not in context, answer with 'unanswerable'. Language: spanish, acc: 86.22%, prompt: Briefly describes the situation and provides the corresponding response. If the answer cannot be found, answer with 'unanswerable'. Language: spanish, acc: 87.05%, prompt: Given the information given, what is the most appropriate response? If the answer cannot be determined, answer with 'unanswerable'. Language: spanish, acc: 87.05%, prompt: Read the following text and give the most accurate answer. If you can't find the answer, answer with 'unanswerable'. Language: spanish, acc: 86.55%, 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: 87.22%, prompt: From the context provided, which response is the most appropriate? If the answer cannot be found, answer with 'unanswerable'. Language: spanish, acc: 87.05%, 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: 86.55%, prompt: According to the information presented, what is the most precise answer? If the answer cannot be determined, answer with 'unanswerable'. Language: spanish, acc: 87.05%, 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: 87.22%, prompt: Based on the context, it provides the most appropriate response. If the answer is not in context, answer with 'unanswerable'. Language: japanese, acc: 87.05%, prompt: Provide the most accurate answer from this context. If the answer isn't in the context, answer 'unanswerable'. Language: japanese, acc: 87.55%, 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: 86.55%, 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: 87.05%, prompt: Provide the most detailed answer based on the given context. If the answer is not in the context, answer 'unanswerable'. Language: japanese, acc: 86.55%, 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: 87.15%, prompt: Based on this context, please provide the most appropriate answer. If the answer is not in the context, answer 'unanswerable'. Language: japanese, acc: 86.91%, 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: 87.05%, 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: 86.65%, 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: 87.05%, prompt: Provide the most detailed answer based on this context. If the answer is not in the context, answer 'unanswerable'. Language: korean, acc: 87.05%, prompt: Give the most accurate answer in context. If the answer is not in context, respond with 'unanswerable'. Language: korean, acc: 86.92%, prompt: Please provide additional information about the facts mentioned in this sentence. If no information is available, respond with 'unanswerable'. Language: korean, acc: 87.05%, 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: 86.92%, 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: 86.32%, prompt: Tell me what you're comparing to in this sentence. If nothing is compared, please respond with 'unanswerable'. Language: korean, acc: 86.82%, 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: 86.55%, prompt: Figure out what information this sentence contains. If no information is available, respond with 'unanswerable'. Language: korean, acc: 87.32%, 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: 87.32%, prompt: Please give the cause of the incident mentioned in the context. If the cause is not clear, respond with 'unanswerable'. Language: korean, acc: 87.15%, prompt: Give expected results in the following sentences. If the result is unpredictable, respond with 'unanswerable'. # iwslt ## 10 prompts Acc: 0.22%, prompt: Convert the following sentence from its original {} language to the target language {}. Acc: 0.22%, prompt: Using machine translation, convert the given sentence from {} into the {} language. Acc: 0.22%, prompt: Please perform a translation of the given sentence, converting it from {} to {}. Acc: 0.22%, prompt: Accurately translate the sentence from {} to {}, ensuring the meaning remains intact. 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 provided sentence from {} to {} while maintaining the original meaning and context. Acc: 0.22%, prompt: Translate the following text from the source language {} to the target language {}. Acc: 0.21%, prompt: Translate the subsequent text passage from its original {} language to the {} language. Acc: 0.21%, prompt: Given the sentence below, perform a machine translation 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: In the role of a language expert, perform a machine translation for the provided sentence, changing 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: Assuming the responsibilities of a professional translator, translate the subsequent text passage from {} to {}. Acc: 0.22%, prompt: As a language specialist, accurately translate the provided 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: Assuming the role of an expert language translator, translate the given text from {} to {}. Acc: 0.22%, prompt: Acting as a professional translator, translate the provided sentence from {} to {}. Acc: 0.22%, prompt: As a skilled linguist, please convert the following sentence from {} to {}. Acc: 0.22%, prompt: While acting as an experienced language expert, translate the following text from {} to {}. ## bertattack Original prompt: Convert the following sentence from its original {} language to the target language {}. Attacked prompt: b'replace the this verb from its originally {} language to the target language {}. ' Original acc: 0.22%, attacked acc: 0.21%, dropped acc: 0.02% Original prompt: Using machine translation, convert the given sentence from {} into the {} language. Attacked prompt: b'on language translation, convert the given sentence from {} into the {} program. ' Original acc: 0.22%, attacked acc: 0.21%, dropped acc: 0.01% Original prompt: Please perform a translation of the given sentence, converting it from {} to {}. Attacked prompt: b'Please take a regular of the given sentence, converting it 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'form a qualified translator, manipulate the given example from its original {} languages to the target language {}. ' Original acc: 0.22%, attacked acc: 0.21%, 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'know the role of a it expert, find a machine translation for the provided sentence, changing it from {} to {}. ' 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 only, please amend the following sentence from {} to {}. ' Original acc: 0.22%, attacked acc: 0.21%, dropped acc: 0.01% ## checklist 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 LkF0FZxMZ4 {}. ' Original acc: 0.22%, attacked acc: 0.22%, 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 AMsRIKZniY. ' Original acc: 0.22%, attacked acc: 0.22%, dropped acc: 0.00% Original prompt: Please perform a translation of the given sentence, converting it from {} to {}. Attacked prompt: b'Please perform a translation of the given sentence, converting it from {} to zexKUTWJYG {}. ' 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 mzjVogkRhM {}. ' Original acc: 0.22%, attacked acc: 0.22%, dropped acc: 0.00% 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 MQnugHcaoy {}. ' 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 MQnugHcaoy {}. ' Original acc: 0.22%, attacked acc: 0.22%, dropped acc: 0.00% ## deepwordbug Original prompt: Convert the following sentence from its original {} language to the target language {}. Attacked prompt: b'tonvert the following senvtence from its original {} language to the target language {}. ' 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, conevrt the gihven Ysentence from {} into the {} languaxe. ' Original acc: 0.22%, attacked acc: 0.21%, dropped acc: 0.01% Original prompt: Please perform a translation of the given sentence, converting it from {} to {}. Attacked prompt: b'Please perform a translation of the given senence, converting it from {} to {}. ' 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 qualiied translator, conevrt the given sentenZe from its oriHinal {} language to the target language {}. ' Original acc: 0.22%, attacked acc: 0.21%, 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 langauge expert, perforF a machine kranslation for the pLovided senteSce, changing it from {} to {}. ' Original acc: 0.22%, attacked acc: 0.14%, dropped acc: 0.08% Original prompt: In the capacity of a proficient translator, please change the following sentence from {} to {}. Attacked prompt: b'In the capaciy of a proficient txanslator, please change the following sentence from {} to {}. ' Original acc: 0.22%, attacked acc: 0.21%, dropped acc: 0.01% ## stresstest 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 true is true {}. ' Original acc: 0.22%, attacked acc: 0.22%, dropped acc: 0.00% 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: 0.22%, dropped acc: 0.00% Original prompt: Please perform a translation of the given sentence, converting it from {} to {}. Attacked prompt: b'Please perform a translation of the given sentence, converting it 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: 0.22%, dropped acc: 0.00% 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: 0.22%, 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 false is not true {}. ' Original acc: 0.22%, attacked acc: 0.22%, dropped acc: 0.01% ## textbugger Original prompt: Convert the following sentence from its original {} language to the target language {}. Attacked prompt: b'Convret the following sentence from its original {} language to the ta\xe2\xb2\x85get language {}. ' 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 equipment translation, convert the given sentence from {} into the {} lan\xc9\xa1uage. ' Original acc: 0.22%, attacked acc: 0.21%, dropped acc: 0.01% Original prompt: Please perform a translation of the given sentence, converting it from {} to {}. Attacked prompt: b'Please fulfilling a t\xe2\xb2\x85anslation of the given sentence, converting it 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'Since a qualifid translator, co\xd5\xb8vert the given sentence from its o riginal {} language to the target language {}. ' Original acc: 0.22%, attacked acc: 0.21%, 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'At the roles of a language expert, fulfilling a m\xc9\x91chine translati\xd0\xben for the provided sentence, changing it from {} to {}. ' 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, pl\xd0\xb5ase \xcf\xb2hange the follow ing sentence from {} to {}. ' Original acc: 0.22%, attacked acc: 0.22%, dropped acc: 0.01% ## textfooler Original prompt: Convert the following sentence from its original {} language to the target language {}. Attacked prompt: b'Conversion the later conviction from its indigenous {} vocabulary to the blanco tongues {}. ' Original acc: 0.22%, attacked acc: 0.13%, dropped acc: 0.09% Original prompt: Using machine translation, convert the given sentence from {} into the {} language. Attacked prompt: b'Using machine translating, transforms the given judgments from {} into the {} lingo. ' Original acc: 0.22%, attacked acc: 0.21%, dropped acc: 0.01% Original prompt: Please perform a translation of the given sentence, converting it from {} to {}. Attacked prompt: b'Please fulfils a rereading of the given conviction, transforms it 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 skilled translation, converting the given sentence from its first {} language to the blanco languages {}. ' Original acc: 0.22%, attacked acc: 0.17%, dropped acc: 0.05% 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 languages specializing, exerting a machine converting for the handed pity, metamorphosis it from {} to {}. ' Original acc: 0.22%, attacked acc: 0.16%, dropped acc: 0.06% Original prompt: In the capacity of a proficient translator, please change the following sentence from {} to {}. Attacked prompt: b'Onto the capacity of a capable artist, please adaptations the next sentence from {} to {}. ' Original acc: 0.22%, attacked acc: 0.21%, dropped acc: 0.01% ## translation Language: chinese, acc: 0.21%, prompt: Please translate the given sentence into {} to {}. Language: chinese, acc: 0.22%, prompt: Please translate the following sentences from {} to {}. Language: chinese, acc: 0.21%, 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: 0.22%, prompt: Please translate the sentences given into {} and convert them into {}. Language: chinese, acc: 0.21%, prompt: Please convert the sentences given to {} to {}. Language: chinese, acc: 0.22%, 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: 0.22%, 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: 0.22%, prompt: Please translate the given sentence to {}, then convert it to {}. Language: french, acc: 0.22%, prompt: Please make a translation of the supplied sentence, transforming it from {} to {}. Language: french, acc: 0.22%, 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: 0.21%, 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: 0.22%, 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: 0.22%, prompt: Please translate the given sentence to {}, then convert it to {}, Language: arabic, acc: 0.21%, prompt: Please translate the given sentence, and convert it from {} to {}, Language: arabic, acc: 0.22%, prompt: Please translate to {}, then convert to {}, Language: arabic, acc: 0.22%, prompt: Please convert the given sentence from {} to {}. Language: spanish, acc: 0.22%, 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: 0.22%, 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: 0.22%, prompt: Please translate the sentence provided to {}, and then turn it to {}. Language: spanish, acc: 0.21%, prompt: Please make a translation of the following sentence, converting it from {} to {}. Language: spanish, acc: 0.22%, 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: 0.22%, 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: 0.21%, prompt: Please convert the following sentences from {} to {}. Language: japanese, acc: 0.22%, 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: 0.22%, prompt: Translate the following sentence into {} and convert it to {}. Language: japanese, acc: 0.21%, 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: 0.20%, 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: 0.21%, prompt: Please translate the given sentence into {}, then {}. Language: korean, acc: 0.22%, prompt: Please translate the given sentence from {} to {}. Language: korean, acc: 0.20%, 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: 0.31%, prompt: Translate the provided sentence from {} to {} while maintaining the original meaning and context. Acc: 0.31%, prompt: Convert the following sentence from its original {} language to the target language {}. Acc: 0.31%, prompt: Translate the following text from the source language {} to the target language {}. Acc: 0.31%, prompt: Given the sentence below, perform a machine translation from {} to {}. Acc: 0.31%, prompt: Translate the subsequent sentence from its source language {} into the desired language {}. Acc: 0.31%, prompt: Perform a machine translation for the provided sentence, changing it from {} to {}. Acc: 0.31%, prompt: Please perform a translation of the given sentence, converting it from {} to {}. Acc: 0.31%, prompt: Translate the subsequent text passage from its original {} language to the {} language. Acc: 0.31%, prompt: Accurately translate the sentence from {} to {}, ensuring the meaning remains intact. Acc: 0.31%, prompt: Using machine translation, convert the given sentence from {} into the {} language. Acc: 0.31%, prompt: As a qualified translator, convert the given sentence from its original {} language to the target language {}. Acc: 0.31%, prompt: In the capacity of a proficient translator, please change the following sentence from {} to {}. Acc: 0.31%, prompt: Taking on the role of a certified translator, perform a translation of the given sentence from {} to {}. Acc: 0.31%, prompt: As a language specialist, accurately translate the provided sentence from {} to {}. Acc: 0.31%, prompt: In the role of a language expert, perform a machine translation for the provided sentence, changing it from {} to {}. Acc: 0.31%, prompt: Assuming the role of an expert language translator, translate the given text from {} to {}. Acc: 0.31%, prompt: Assuming the responsibilities of a professional translator, translate the subsequent text passage from {} to {}. Acc: 0.31%, prompt: Acting as a professional translator, translate the provided sentence from {} to {}. Acc: 0.31%, prompt: While acting as an experienced language expert, translate the following text from {} to {}. Acc: 0.30%, prompt: As a skilled linguist, please convert the following sentence from {} to {}. ## bertattack 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 maintained the same meaning and semantics. ' Original acc: 0.31%, attacked acc: 0.31%, dropped acc: 0.00% Original prompt: Convert the following sentence from its original {} language to the target language {}. Attacked prompt: b'translate the second sentence from its original {} language to the target language {}. ' Original acc: 0.31%, attacked acc: 0.30%, dropped acc: 0.00% Original prompt: Translate the following text from the source language {} to the target language {}. Attacked prompt: b'Translate the final text from the source language {} to the target python {}. ' Original acc: 0.31%, attacked acc: 0.30%, 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 trained translator, convert the given case from its original {} language to the project languages {}. ' Original acc: 0.31%, attacked acc: 0.30%, 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 ability of a proficient translator, please visit the following sentence from {} to {}. ' Original acc: 0.31%, attacked acc: 0.31%, 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, for a translate of the the segment from {} to {}. ' Original acc: 0.31%, attacked acc: 0.30%, dropped acc: 0.01% ## checklist 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 ofOQXLxiW4. ' Original acc: 0.31%, attacked acc: 0.31%, dropped acc: 0.00% 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 OB4KVJzIft {}. ' Original acc: 0.31%, attacked acc: 0.31%, dropped acc: 0.00% 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 Cev0s49fbe {}. ' Original acc: 0.31%, attacked acc: 0.30%, 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 qualified translator, convert the given sentence from its original {} language to the target language yh912BU8T0 {}. ' Original acc: 0.31%, attacked acc: 0.31%, 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 OB4KVJzIft {}. ' Original acc: 0.31%, attacked acc: 0.29%, 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'Taking on the role of a certified translator, perform a translation of the given sentence from {} to OB4KVJzIft {}. ' Original acc: 0.31%, attacked acc: 0.30%, dropped acc: 0.01% ## deepwordbug Original prompt: Translate the provided sentence from {} to {} while maintaining the original meaning and context. Attacked prompt: b'Translate the proviWded entence from {} to {} while maintaining the oriinal meanin and conEtext. ' Original acc: 0.31%, attacked acc: 0.30%, dropped acc: 0.01% Original prompt: Convert the following sentence from its original {} language to the target language {}. Attacked prompt: b'oCnvert the following sentence from its original {} language to the target language {}. ' Original acc: 0.31%, attacked acc: 0.31%, dropped acc: 0.00% Original prompt: Translate the following text from the source language {} to the target language {}. Attacked prompt: b'Translate the followiSng text from the souce language {} to the targea Rlanguage {}. ' Original acc: 0.31%, attacked acc: 0.30%, 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'A a qalified translator, convert the given esntence from its original {} language to the trget languOge {}. ' Original acc: 0.31%, attacked acc: 0.30%, 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 capaOity of a proficient translator, please hange the following sentence from {} to {}. ' Original acc: 0.31%, attacked acc: 0.31%, 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 roEle of a certifLed ltranslator, perofrm a trasnlation of the giDven senence from {} to {}. ' Original acc: 0.31%, attacked acc: 0.26%, dropped acc: 0.05% ## 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 true is true and true is true and true is true and true is true and true is true . ' Original acc: 0.31%, attacked acc: 0.31%, dropped acc: -0.00% 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: 0.31%, attacked acc: 0.31%, dropped acc: -0.00% 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 false is not true {}. ' Original acc: 0.31%, attacked acc: 0.30%, 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 true is true {}. ' Original acc: 0.31%, attacked acc: 0.31%, 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 and false is not true {}. ' Original acc: 0.31%, attacked acc: 0.31%, 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 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: 0.31%, attacked acc: 0.30%, 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 provided s\xd0\xb5ntence from {} to {} while maintaining the oriinal meaning and context. ' Original acc: 0.31%, attacked acc: 0.30%, dropped acc: 0.01% Original prompt: Convert the following sentence from its original {} language to the target language {}. Attacked prompt: b'Convrt the following sentence from its original {} language to the objectives language {}. ' Original acc: 0.31%, attacked acc: 0.31%, dropped acc: 0.00% Original prompt: Translate the following text from the source language {} to the target language {}. Attacked prompt: b'Translate the subsequent text from the source tongue {} to the target lanugage {}. ' Original acc: 0.31%, attacked acc: 0.29%, dropped acc: 0.02% 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 punishments from its original {} parlance to the target language {}. ' Original acc: 0.31%, attacked acc: 0.30%, 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, pelase chan\xc9\xa1e the following sentence from {} to {}. ' Original acc: 0.31%, attacked acc: 0.31%, 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 role of a certified translator, performed a translator of the gi\xd1\xb5en sentenc\xd0\xb5 from {} to {}. ' Original acc: 0.31%, attacked acc: 0.28%, dropped acc: 0.04% ## textfooler Original prompt: Translate the provided sentence from {} to {} while maintaining the original meaning and context. Attacked prompt: b'Translate the provided judgments from {} to {} while protection the preliminary meaning and regard. ' Original acc: 0.31%, attacked acc: 0.30%, dropped acc: 0.01% Original prompt: Convert the following sentence from its original {} language to the target language {}. Attacked prompt: b'Convert the below sentence from its indigenous {} language to the blanco language {}. ' Original acc: 0.31%, attacked acc: 0.22%, dropped acc: 0.09% Original prompt: Translate the following text from the source language {} to the target language {}. Attacked prompt: b'Translate the follows text from the source language {} to the purposes wording {}. ' Original acc: 0.31%, attacked acc: 0.30%, 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 trained translator, convert the given verdict from its original {} phraseology to the blanco language {}. ' Original acc: 0.31%, attacked acc: 0.28%, 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 functionality of a specialised interpreting, please amends the subsequent judgments from {} to {}. ' Original acc: 0.31%, attacked acc: 0.30%, 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 attest translator, implement a interpreters of the gave sanction from {} to {}. ' Original acc: 0.31%, attacked acc: 0.26%, dropped acc: 0.05% ## translation Language: chinese, acc: 0.30%, prompt: Please translate the given sentence into {} to {}. Language: chinese, acc: 0.31%, prompt: Please translate the following sentences from {} to {}. Language: chinese, acc: 0.30%, prompt: Please convert the following sentences to {} and translate to {}. Language: chinese, acc: 0.31%, prompt: Please convert the given sentence from {} to {}. Language: chinese, acc: 0.31%, prompt: Please translate the next sentence from {} to {}. Language: chinese, acc: 0.31%, prompt: Please translate the following sentence from {} to {}. Language: chinese, acc: 0.31%, prompt: Please translate the sentences given into {} and convert them into {}. Language: chinese, acc: 0.31%, prompt: Please convert the sentences given to {} to {}. Language: chinese, acc: 0.31%, prompt: Please translate the following sentences into {} and convert them into {}. Language: chinese, acc: 0.31%, prompt: Please change the given sentence from {} to {}. Language: french, acc: 0.31%, prompt: Please translate the given sentence, converting it from {} to {}. Language: french, acc: 0.31%, prompt: Please translate the following sentence from {} to {}. Language: french, acc: 0.30%, prompt: Please turn the sentence below into {}, then translate it into {}. Language: french, acc: 0.31%, prompt: Please convert the given phrase from {} to {}. Language: french, acc: 0.31%, prompt: Please translate the following sentence from {} to {}. Language: french, acc: 0.31%, prompt: Please translate the sentence below from {} to {}. Language: french, acc: 0.31%, prompt: Please translate the given sentence to {}, then convert it to {}. Language: french, acc: 0.31%, prompt: Please make a translation of the supplied sentence, transforming it from {} to {}. Language: french, acc: 0.31%, prompt: Please translate the following sentence to {}, then convert it to {}. Language: french, acc: 0.31%, prompt: Please transform the given sentence from {} to {}. Language: arabic, acc: 0.31%, prompt: Please translate the given sentence, and convert it from {} to {}, Language: arabic, acc: 0.30%, prompt: Please translate the following sentence from {} to {}, Language: arabic, acc: 0.31%, prompt: Please convert the sentence below to {}, and then translate it to {}, Language: arabic, acc: 0.31%, prompt: Please convert the given sentence from {} to {}, Language: arabic, acc: 0.30%, prompt: Please translate the following sentence from {} to {}, Language: arabic, acc: 0.31%, prompt: Please convert the sentence below from {} to {}, Language: arabic, acc: 0.31%, prompt: Please translate the given sentence to {}, then convert it to {}, Language: arabic, acc: 0.31%, prompt: Please translate the given sentence, and convert it from {} to {}, Language: arabic, acc: 0.31%, prompt: Please translate to {}, then convert to {}, Language: arabic, acc: 0.31%, prompt: Please convert the given sentence from {} to {}. Language: spanish, acc: 0.31%, prompt: Please make a translation of the provided phrase, converting it from {} to {}. Language: spanish, acc: 0.31%, prompt: Please translate the following sentence from {} to {}. Language: spanish, acc: 0.30%, prompt: Please convert the next sentence to {}, and then translate it to {}. Language: spanish, acc: 0.31%, prompt: Please make a translation of the given phrase, converting it from {} to {}. Language: spanish, acc: 0.31%, prompt: Please translate the following sentence from {} to {}. Language: spanish, acc: 0.31%, prompt: Please convert the following sentence from {} to {}. Language: spanish, acc: 0.30%, prompt: Please translate the sentence provided to {}, and then turn it to {}. Language: spanish, acc: 0.31%, prompt: Please make a translation of the following sentence, converting it from {} to {}. Language: spanish, acc: 0.30%, prompt: Please translate the next sentence to {}, and then turn it to {}. Language: spanish, acc: 0.31%, prompt: Please convert the given sentence from {} to {}. Language: japanese, acc: 0.31%, prompt: Please translate the given sentence from {} to {}. Language: japanese, acc: 0.31%, prompt: Please translate the following sentence from {} to {}. Language: japanese, acc: 0.30%, prompt: Please convert the following sentences into {} and translate them into {}. Language: japanese, acc: 0.31%, prompt: Please translate the given sentence by converting {} to {}. Language: japanese, acc: 0.31%, prompt: Please translate the following sentence from {} to {}. Language: japanese, acc: 0.31%, prompt: Please convert the following sentences from {} to {}. Language: japanese, acc: 0.31%, prompt: Translate the given sentence into {} and convert it to {}. Language: japanese, acc: 0.31%, prompt: Please translate the given sentence from {} to {}. Language: japanese, acc: 0.31%, prompt: Translate the following sentence into {} and convert it to {}. Language: japanese, acc: 0.31%, prompt: Convert the given statement from {} to {}. Language: korean, acc: 0.31%, prompt: Please translate the given sentence from {} to {}. Language: korean, acc: 0.31%, prompt: Please translate the following sentence from {} to {}. Language: korean, acc: 0.31%, prompt: Please translate the sentences below into {}, then {}. Language: korean, acc: 0.31%, prompt: Please translate the given sentences from {} to {}. Language: korean, acc: 0.31%, prompt: Please translate the following sentence from {} to {}. Language: korean, acc: 0.31%, prompt: Please convert the sentences below from {} to {}. Language: korean, acc: 0.31%, prompt: Please translate the given sentence into {}, then {}. Language: korean, acc: 0.31%, prompt: Please translate the given sentence from {} to {}. Language: korean, acc: 0.31%, prompt: Please translate the following sentences into {}, then {}. Language: korean, acc: 0.31%, prompt: Please convert the given sentence from {} to {}. # math ## 10 prompts Acc: 17.50%, prompt: Find the solution for this mathematical challenge with {}. Acc: 17.50%, prompt: Resolve the following mathematical question associated with {}. Acc: 17.50%, prompt: Figure out the solution for the following mathematical task with {}. Acc: 17.50%, prompt: Evaluate the following mathematical problem that includes {}. Acc: 16.88%, prompt: Determine the solution to this mathematical problem related to {}. Acc: 16.88%, prompt: Calculate the answer to the following math query about {}. Acc: 16.25%, prompt: Solve the following math question about {}. Acc: 16.25%, prompt: Compute the result of this math task concerning {}. Acc: 16.25%, prompt: Work out the answer to this math problem featuring {}. Acc: 16.25%, prompt: Obtain the result for this math question regarding {}. Acc: 18.12%, prompt: Embodying a mathematics advisor, find the answer to the following math problem with {}. Acc: 17.50%, prompt: As a mathematics instructor, calculate the answer to the following problem related to {}. Acc: 17.50%, prompt: Imagine being a math mentor and determine the solution to the following problem with {}. Acc: 16.88%, prompt: Pretend you are a math teacher and solve this mathematical question concerning {}. Acc: 16.88%, prompt: In the position of a mathematics coach, figure out the answer to this math challenge involving {}. Acc: 16.88%, prompt: As a mathematics consultant, resolve the following problem associated with {}. Acc: 16.25%, prompt: As a math guru, evaluate the following mathematical task regarding {}. Acc: 16.25%, prompt: In the shoes of a mathematics master, obtain the result for this math question that includes {}. Acc: 15.62%, prompt: Act as a math expert and compute the result of the following mathematical task about {}. Acc: 15.00%, prompt: Taking on the role of a math specialist, work out the solution for this mathematical question featuring {}. ## bertattack Original prompt: Find the solution for this mathematical challenge with {}. Attacked prompt: b'Find the conclusion for this mathematical challenge with {}. ' Original acc: 17.50%, attacked acc: 15.00%, dropped acc: 2.50% Original prompt: Resolve the following mathematical question associated with {}. Attacked prompt: b'about the following mathematical question associated with {}. ' Original acc: 17.50%, attacked acc: 15.62%, dropped acc: 1.87% Original prompt: Figure out the solution for the following mathematical task with {}. Attacked prompt: b'Figure out the sum for the following mathematical task with {}. ' Original acc: 17.50%, attacked acc: 15.62%, dropped acc: 1.87% Original prompt: Embodying a mathematics advisor, find the answer to the following math problem with {}. Attacked prompt: b'Embodying a mathematics advisor, find the subject to the such math item with {}. ' Original acc: 18.12%, attacked acc: 9.38%, dropped acc: 8.75% Original prompt: As a mathematics instructor, calculate the answer to the following problem related to {}. Attacked prompt: b'As a mathematics instructor, calculate the inverse to the following problem by to {}. ' Original acc: 17.50%, attacked acc: 14.37%, dropped acc: 3.12% Original prompt: Imagine being a math mentor and determine the solution to the following problem with {}. Attacked prompt: b'[UNK] being a math mentor and determine the solve to the following problem with {}. ' Original acc: 17.50%, attacked acc: 15.62%, dropped acc: 1.87% ## checklist Original prompt: Find the solution for this mathematical challenge with {}. Attacked prompt: b'Find the solution for this mathematical challenge with 9Q3S1F94fE {}. ' Original acc: 17.50%, attacked acc: 16.25%, dropped acc: 1.25% Original prompt: Resolve the following mathematical question associated with {}. Attacked prompt: b'Resolve the following mathematical question associated with 6a4Yn3RGVc {}. ' Original acc: 17.50%, attacked acc: 15.62%, dropped acc: 1.87% Original prompt: Figure out the solution for the following mathematical task with {}. Attacked prompt: b'Figure out the solution for the following mathematical task with 6a4Yn3RGVc {}. ' Original acc: 17.50%, attacked acc: 16.25%, 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, find the answer to the following math problem with EAB4KP2NVY {}. ' Original acc: 18.12%, attacked acc: 16.25%, dropped acc: 1.87% 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 5yWbBXztUY {}. ' Original acc: 17.50%, attacked acc: 16.88%, dropped acc: 0.62% Original prompt: Imagine being a math mentor and determine the solution to the following problem with {}. Attacked prompt: b'Imagine being a math mentor and determine the solution to the following problem with b74X5IVXQY {}. ' Original acc: 17.50%, attacked acc: 15.62%, dropped acc: 1.87% ## deepwordbug Original prompt: Find the solution for this mathematical challenge with {}. Attacked prompt: b'Find the aolution for this amthematical challenge with {}. ' Original acc: 17.50%, attacked acc: 16.25%, dropped acc: 1.25% Original prompt: Resolve the following mathematical question associated with {}. Attacked prompt: b'xesolve the following mathematicMal question associated with {}. ' Original acc: 17.50%, attacked acc: 16.25%, dropped acc: 1.25% Original prompt: Figure out the solution for the following mathematical task with {}. Attacked prompt: b'Figure out the slution for the following mathemaitcal task with {}. ' Original acc: 17.50%, attacked acc: 15.62%, dropped acc: 1.87% 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 prblem with {}. ' Original acc: 18.12%, attacked acc: 16.25%, dropped acc: 1.87% Original prompt: As a mathematics instructor, calculate the answer to the following problem related to {}. Attacked prompt: b'sA a mathematics instructor, calculate the anvwer to the following problem relDated to {}. ' Original acc: 17.50%, attacked acc: 15.00%, dropped acc: 2.50% Original prompt: Imagine being a math mentor and determine the solution to the following problem with {}. Attacked prompt: b'Imagine being a math mentor and determine the solution to the following problem with {}. ' Original acc: 17.50%, attacked acc: 17.50%, dropped acc: 0.00% ## stresstest Original prompt: Find the solution for this mathematical challenge with {}. Attacked prompt: b'Find the solution for this mathematical challenge with and false is not true {}. ' Original acc: 17.50%, attacked acc: 16.25%, dropped acc: 1.25% Original prompt: Resolve the following mathematical question associated with {}. Attacked prompt: b'Resolve the following mathematical question associated with and false is not true {}. ' Original acc: 17.50%, attacked acc: 16.25%, dropped acc: 1.25% Original prompt: Figure out the solution for the following mathematical task with {}. Attacked prompt: b'Figure out the solution for the following mathematical task with and false is not true {}. ' Original acc: 17.50%, attacked acc: 16.88%, dropped acc: 0.62% 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 false is not true {}. ' Original acc: 18.12%, attacked acc: 16.25%, dropped acc: 1.87% 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 false is not true {}. ' Original acc: 17.50%, attacked acc: 16.25%, dropped acc: 1.25% Original prompt: Imagine being a math mentor and determine the solution to the following problem with {}. Attacked prompt: b'Imagine being a math mentor and determine the solution to the following problem with and false is not true {}. ' Original acc: 17.50%, attacked acc: 16.25%, dropped acc: 1.25% ## textbugger Original prompt: Find the solution for this mathematical challenge with {}. Attacked prompt: b'Find the solve for this mathematical challenge with {}. ' Original acc: 17.50%, attacked acc: 16.88%, dropped acc: 0.62% Original prompt: Resolve the following mathematical question associated with {}. Attacked prompt: b'Reslove the following math question associated with {}. ' Original acc: 17.50%, attacked acc: 15.62%, dropped acc: 1.87% Original prompt: Figure out the solution for the following mathematical task with {}. Attacked prompt: b'Figure out the solution for the following arithmetic task with {}. ' Original acc: 17.50%, attacked acc: 16.88%, dropped acc: 0.62% Original prompt: Embodying a mathematics advisor, find the answer to the following math problem with {}. Attacked prompt: b'Embodying a mathematics advis\xd0\xber, find the ansewr to the following math problem with {}. ' Original acc: 18.12%, attacked acc: 16.25%, dropped acc: 1.87% Original prompt: As a mathematics instructor, calculate the answer to the following problem related to {}. Attacked prompt: b'As a mathematics trainers, calculate the respond to the following problem related to {}. ' Original acc: 17.50%, attacked acc: 15.62%, dropped acc: 1.87% Original prompt: Imagine being a math mentor and determine the solution to the following problem with {}. Attacked prompt: b'Ima\xc9\xa1ine being a math mentor and determi\xd5\xb8e the solution to the following problem with {}. ' Original acc: 17.50%, attacked acc: 16.25%, dropped acc: 1.25% ## textfooler Original prompt: Find the solution for this mathematical challenge with {}. Attacked prompt: b'Find the resolutions for this mathematical challenge with {}. ' Original acc: 17.50%, attacked acc: 15.00%, dropped acc: 2.50% Original prompt: Resolve the following mathematical question associated with {}. Attacked prompt: b'Liquidate the following mathematical question affiliates with {}. ' Original acc: 17.50%, attacked acc: 15.00%, dropped acc: 2.50% Original prompt: Figure out the solution for the following mathematical task with {}. Attacked prompt: b'Fig out the resolution for the following arithmetic task with {}. ' Original acc: 17.50%, attacked acc: 15.62%, dropped acc: 1.87% Original prompt: Embodying a mathematics advisor, find the answer to the following math problem with {}. Attacked prompt: b'Embodying a mathematics advice, find the rebuttal to the following math problem with {}. ' Original acc: 18.12%, attacked acc: 13.75%, dropped acc: 4.37% Original prompt: As a mathematics instructor, calculate the answer to the following problem related to {}. Attacked prompt: b'As a mathematics trainers, calculate the answered to the following problem related to {}. ' Original acc: 17.50%, attacked acc: 15.00%, dropped acc: 2.50% Original prompt: Imagine being a math mentor and determine the solution to the following problem with {}. Attacked prompt: b'Imagine being a math mentor and determine the solution to the following question with {}. ' Original acc: 17.50%, attacked acc: 16.25%, dropped acc: 1.25% ## translation Language: chinese, acc: 17.50%, prompt: Calculate the answer to a mathematical question about {}. Language: chinese, acc: 16.88%, prompt: Calculate the value of {}. Language: chinese, acc: 17.50%, prompt: Please calculate the result of {}. Language: chinese, acc: 18.12%, prompt: According to {}, solve the following mathematical problem. Language: chinese, acc: 17.50%, prompt: Do the math with {}. Language: chinese, acc: 18.12%, prompt: Solve mathematical problems involving {}. Language: chinese, acc: 17.50%, prompt: Please calculate the answer involving {}. Language: chinese, acc: 17.50%, prompt: Solve for the following values based on {}. Language: chinese, acc: 17.50%, prompt: Calculate the following mathematical tasks using {}. Language: chinese, acc: 17.50%, prompt: Calculate the answer to the {} related question. Language: french, acc: 16.88%, prompt: Calculate the answer to the following mathematical question concerning {}. Language: french, acc: 17.50%, prompt: Calculate the result of {}. Language: french, acc: 17.50%, prompt: Please calculate the value of {}. Language: french, acc: 18.12%, prompt: According to {}, solve the following mathematical problem. Language: french, acc: 16.88%, prompt: Perform mathematical calculations with {}. Language: french, acc: 17.50%, prompt: Solve the mathematical problem involving {}. Language: french, acc: 16.88%, prompt: Please calculate the answer related to {}. Language: french, acc: 15.62%, prompt: According to {}, set the following value. Language: french, acc: 17.50%, prompt: Perform the following mathematical task using {}. Language: french, acc: 16.88%, prompt: Calculate the answer to the questions related to {}. Language: arabic, acc: 16.88%, prompt: Compute the answer to the next mathematical question about {}. Language: arabic, acc: 16.25%, prompt: Calculate {}. Language: arabic, acc: 17.50%, prompt: Please calculate {}. Language: arabic, acc: 18.12%, prompt: According to {}, solve the following mathematical problem. Language: arabic, acc: 16.25%, prompt: Do mathematical calculations using {}. Language: arabic, acc: 16.25%, prompt: A solution to the mathematical problem involving {}. Language: arabic, acc: 16.88%, prompt: Please calculate the answer regarding {}. Language: arabic, acc: 15.62%, prompt: According to {}, determine the next value. Language: arabic, acc: 18.12%, prompt: DO THE NEXT MATHEMATICAL JOB USING {}. Language: arabic, acc: 17.50%, prompt: Calculate the answer to questions related to {}. Language: spanish, acc: 16.88%, prompt: Compute the answer to the following mathematical question on {}. Language: spanish, acc: 16.88%, prompt: Compute the result of {}. Language: spanish, acc: 17.50%, prompt: Please calculate the value of {}. Language: spanish, acc: 17.50%, prompt: As {}, it solves the following mathematical problem. Language: spanish, acc: 16.88%, prompt: Performs mathematical calculations using {}. Language: spanish, acc: 17.50%, prompt: Solve the mathematical problem involving {}. Language: spanish, acc: 16.88%, prompt: Please calculate the answer related to {}. Language: spanish, acc: 15.62%, prompt: As {}, determine the next value. Language: spanish, acc: 17.50%, prompt: Perform the following mathematical task using {}. Language: spanish, acc: 16.25%, prompt: Compute the answer to questions related to {}. Language: japanese, acc: 16.25%, prompt: Calculate the answers to the math questions about {}. Language: japanese, acc: 16.88%, prompt: Calculate the value of {}. Language: japanese, acc: 16.25%, prompt: Please find the answer to {}. Language: japanese, acc: 18.12%, prompt: Based on {}, please solve the following mathematical problems. Language: japanese, acc: 16.88%, prompt: Use {} to perform mathematical calculations. Language: japanese, acc: 15.62%, prompt: Please solve the math problem that contains {}. Language: japanese, acc: 16.88%, prompt: Please calculate the answers related to {}. Language: japanese, acc: 17.50%, prompt: Based on {}, find the following values: Language: japanese, acc: 17.50%, prompt: Use {} to solve the following mathematical problem. Language: japanese, acc: 16.88%, prompt: Please calculate the answers to the questions related to {}. Language: korean, acc: 16.88%, prompt: Calculate the answer of the following math problem to {}. Language: korean, acc: 17.50%, prompt: Calculate the result of {}. Language: korean, acc: 17.50%, prompt: Please calculate the value of {}. Language: korean, acc: 16.88%, prompt: Work out the following math problems according to {}. Language: korean, acc: 16.88%, prompt: Use {} to proceed with mathematical calculations. Language: korean, acc: 17.50%, prompt: Work out a math problem involving {}. Language: korean, acc: 16.88%, prompt: Please calculate the answer to {}. Language: korean, acc: 16.25%, prompt: Try to get the following values according to {}. Language: korean, acc: 16.88%, prompt: Work out the next math task using {}. Language: korean, acc: 16.88%, prompt: Calculate the answer of the problem involving {}.