[ { "dataset": "cola", "type": "task", "origin_acc": 78.1, "origin_prompt": "Assess the following sentence and determine if it is grammatically correct. Respond with 'Acceptable' or 'Unacceptable'.", "atk_acc": 77.0, "atk_prompt": "Assess the following sentence and determine if it is grammatically correct. Respond with 'Acceptable' or 'Unacceptable' zexKUTWJYG. ", "transfer_ori_acc": 80.0, "transfer_atk_acc": 79.5 }, { "dataset": "cola", "type": "role", "origin_acc": 75.7, "origin_prompt": "While performing grammar analysis, classify the grammar of the following sentence as 'acceptable' for correct grammar or 'unacceptable' for incorrect grammar.", "atk_acc": 74.6, "atk_prompt": "While performing grammar analysis, classify the grammar of the following sentence as 'acceptable' for correct grammar or 'unacceptable' for incorrect grammar zexKUTWJYG. ", "transfer_ori_acc": 82.0, "transfer_atk_acc": 81.0 }, { "dataset": "mrpc", "type": "task", "origin_acc": 82.84, "origin_prompt": "Please analyze the provided sentences and indicate if their meanings are the same by choosing 'equivalent' or 'not_equivalent'.", "atk_acc": 80.88, "atk_prompt": "Please analyze the provided sentences and indicate if their meanings are the same by choosing 'equivalent' or 'not_equivalent' CuVpbbkC4c. ", "transfer_ori_acc": 71.0, "transfer_atk_acc": 70.5 }, { "dataset": "mrpc", "type": "role", "origin_acc": 80.15, "origin_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'.", "atk_acc": 79.66, "atk_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' BWmsLx8xOA. ", "transfer_ori_acc": 71.5, "transfer_atk_acc": 70.5 }, { "dataset": "qnli", "type": "task", "origin_acc": 87.8, "origin_prompt": "Based on the information in the context, decide if the answer to the question is justified by choosing 'entailment' or 'not_entailment'.", "atk_acc": 86.4, "atk_prompt": "Based on the information in the context, decide if the answer to the question is justified by choosing 'entailment' or 'not_entailment' mOJ9nEwT4f. ", "transfer_ori_acc": 82.0, "transfer_atk_acc": 80.5 }, { "dataset": "qnli", "type": "role", "origin_acc": 88.5, "origin_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'.", "atk_acc": 87.7, "atk_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' MSAddJ4D2a. ", "transfer_ori_acc": 81.5, "transfer_atk_acc": 83.0 }, { "dataset": "qqp", "type": "task", "origin_acc": 87.6, "origin_prompt": "Are the meanings of these two phrases the same? Reply with 'equivalent' or 'not_equivalent'.", "atk_acc": 87.4, "atk_prompt": "Are the meanings of these two phrases the same? Reply with 'equivalent' or 'not_equivalent' Pbl3tYuoRo. ", "transfer_ori_acc": 81.0, "transfer_atk_acc": 79.5 }, { "dataset": "qqp", "type": "role", "origin_acc": 86.5, "origin_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.", "atk_acc": 86.0, "atk_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 6a4Yn3RGVc. ", "transfer_ori_acc": 83.0, "transfer_atk_acc": 83.0 }, { "dataset": "rte", "type": "task", "origin_acc": 84.84, "origin_prompt": "Identify whether the given pair of sentences demonstrates entailment or not_entailment. Answer with 'entailment' or 'not_entailment'.", "atk_acc": 83.39, "atk_prompt": "Identify whether the given pair of sentences demonstrates entailment or not_entailment. Answer with 'entailment' or 'not_entailment' 6a4Yn3RGVc. ", "transfer_ori_acc": 82.0, "transfer_atk_acc": 82.5 }, { "dataset": "rte", "type": "role", "origin_acc": 84.12, "origin_prompt": "In the capacity of an entailment assessment system, indicate if the link between the following sentences is 'entailment' or 'not_entailment'.", "atk_acc": 82.31, "atk_prompt": "In the capacity of an entailment assessment system, indicate if the link between the following sentences is 'entailment' or 'not_entailment' MQnugHcaoy. ", "transfer_ori_acc": 82.0, "transfer_atk_acc": 81.5 }, { "dataset": "sst2", "type": "task", "origin_acc": 94.95, "origin_prompt": "Analyze the tone of this statement and respond with either 'positive' or 'negative'.", "atk_acc": 94.5, "atk_prompt": "Analyze the tone of this statement and respond with either 'positive' or 'negative' XStsD2pUzu. ", "transfer_ori_acc": 94.5, "transfer_atk_acc": 97.0 }, { "dataset": "sst2", "type": "role", "origin_acc": 95.41, "origin_prompt": "Acting as a sentiment evaluator, identify if the given sentence is 'positive' or 'negative'.", "atk_acc": 94.84, "atk_prompt": "Acting as a sentiment evaluator, identify if the given sentence is 'positive' or 'negative' SFPCqpiClT. ", "transfer_ori_acc": 96.5, "transfer_atk_acc": 96.0 }, { "dataset": "wnli", "type": "task", "origin_acc": 64.79, "origin_prompt": "Assess the connection between the following sentences and classify it as 'entailment' or 'not_entailment'.", "atk_acc": 63.38, "atk_prompt": "Assess the connection between the following sentences and classify it as 'entailment' or 'not_entailment' Cev0s49fbe. ", "transfer_ori_acc": 66.19718309859155, "transfer_atk_acc": 69.01408450704226 }, { "dataset": "wnli", "type": "role", "origin_acc": 66.2, "origin_prompt": "As an entailment identification system, examine the connection between the following sentences and respond with 'entailment' or 'not_entailment'.", "atk_acc": 66.2, "atk_prompt": "As an entailment identification system, examine the connection between the following sentences and respond with 'entailment' or 'not_entailment' 5XeflW1ZJc. ", "transfer_ori_acc": 60.56338028169014, "transfer_atk_acc": 63.38028169014085 }, { "dataset": "mmlu", "type": "task", "origin_acc": 45.92, "origin_prompt": "Answer the subsequent multiple-choice question about {} by picking the right option among 'A', 'B', 'C', or 'D'.", "atk_acc": 45.04, "atk_prompt": "Answer the subsequent multiple-choice question about {} by picking the right option among 'A', 'B', 'C', or 'D' KjPJJ2a7RB. ", "transfer_ori_acc": 66.0, "transfer_atk_acc": 66.0 }, { "dataset": "mmlu", "type": "role", "origin_acc": 46.99, "origin_prompt": "Considering your familiarity with {}, attend to the following multiple-choice question by picking 'A', 'B', 'C', or 'D'.", "atk_acc": 45.04, "atk_prompt": "Considering your familiarity with {}, attend to the following multiple-choice question by picking 'A', 'B', 'C', or 'D' SFPCqpiClT.", "transfer_ori_acc": 65.5, "transfer_atk_acc": 65.5 }, { "dataset": "squad_v2", "type": "task", "origin_acc": 87.91, "origin_prompt": "From the context, provide the most precise answer. If the answer is not in the context, respond with 'unanswerable'.", "atk_acc": 87.05, "atk_prompt": "From the context, provide the most precise answer. If the answer is not in the context, respond with 'unanswerable' 5yWbBXztUY.", "transfer_ori_acc": 69.33640054239395, "transfer_atk_acc": 68.59989260588603 }, { "dataset": "squad_v2", "type": "role", "origin_acc": 87.05, "origin_prompt": "Drawing upon your expertise in the context, determine the most suitable answer. If an answer isn't available, state 'unanswerable'.", "atk_acc": 86.55, "atk_prompt": "Drawing upon your expertise in the context, determine the most suitable answer. If an answer isn't available, state 'unanswerable' 5XeflW1ZJc.", "transfer_ori_acc": 71.1631951530121, "transfer_atk_acc": 74.10685231284573 }, { "dataset": "iwslt", "type": "task", "origin_acc": 0.22, "origin_prompt": "Convert the following sentence from its original {} language to the target language {}.", "atk_acc": 0.22, "atk_prompt": "Convert the following sentence from its original {} language to the target language LkF0FZxMZ4 {}. ", "transfer_ori_acc": 0.23991880278182126, "transfer_atk_acc": 0.2348130480000894 }, { "dataset": "iwslt", "type": "role", "origin_acc": 0.22, "origin_prompt": "In the role of a language expert, perform a machine translation for the provided sentence, changing it from {} to {}.", "atk_acc": 0.21, "atk_prompt": "In the role of a language expert, perform a machine translation for the provided sentence, changing it from {} to MQnugHcaoy {}. ", "transfer_ori_acc": 0.244428238867859, "transfer_atk_acc": 0.2280062687297454 }, { "dataset": "un_multi", "type": "task", "origin_acc": 0.31, "origin_prompt": "Translate the following text from the source language {} to the target language {}.", "atk_acc": 0.3, "atk_prompt": "Translate the following text from the source language {} to the target language Cev0s49fbe {}. ", "transfer_ori_acc": 0.3520628408924414, "transfer_atk_acc": 0.35416969837419654 }, { "dataset": "un_multi", "type": "role", "origin_acc": 0.31, "origin_prompt": "In the capacity of a proficient translator, please change the following sentence from {} to {}.", "atk_acc": 0.29, "atk_prompt": "In the capacity of a proficient translator, please change the following sentence from {} to OB4KVJzIft {}. ", "transfer_ori_acc": 0.3486649402303781, "transfer_atk_acc": 0.3424551550058589 }, { "dataset": "math", "type": "task", "origin_acc": 17.5, "origin_prompt": "Resolve the following mathematical question associated with {}.", "atk_acc": 15.62, "atk_prompt": "Resolve the following mathematical question associated with 6a4Yn3RGVc {}. ", "transfer_ori_acc": 22.5, "transfer_atk_acc": 24.375 }, { "dataset": "math", "type": "role", "origin_acc": 17.5, "origin_prompt": "Imagine being a math mentor and determine the solution to the following problem with {}.", "atk_acc": 15.62, "atk_prompt": "Imagine being a math mentor and determine the solution to the following problem with b74X5IVXQY {}. ", "transfer_ori_acc": 22.5, "transfer_atk_acc": 23.75 } ]