[ { "dataset": "cola", "type": "task", "origin_acc": 77.9, "origin_prompt": "Examine the given sentence and decide if it is grammatically sound. Answer with either 'Acceptable' or 'Unacceptable':", "atk_acc": 62.6, "atk_prompt": "Examine the given sentence and decide if it is grammatically sound. Answer with either 'Acceptable' or 'Unacceptable' IhuBIhoPGc:", "transfer_ori_acc": 50.0, "transfer_atk_acc": 50.6 }, { "dataset": "cola", "type": "role", "origin_acc": 76.3, "origin_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:", "atk_acc": 74.9, "atk_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 IhuBIhoPGc:", "transfer_ori_acc": 36.199999999999996, "transfer_atk_acc": 30.599999999999998 }, { "dataset": "mrpc", "type": "task", "origin_acc": 82.11, "origin_prompt": "Do these two sentences have the same underlying meaning? Respond with 'equivalent' or 'not_equivalent'.", "atk_acc": 79.17, "atk_prompt": "Do these two sentences have the same underlying meaning? Respond with 'equivalent' or 'not_equivalent' MSAddJ4D2a. ", "transfer_ori_acc": 54.90196078431373, "transfer_atk_acc": 39.705882352941174 }, { "dataset": "mrpc", "type": "role", "origin_acc": 82.6, "origin_prompt": "As a semantic comparison expert, evaluate the given pair of sentences and determine if they are 'equivalent' or 'not_equivalent'.", "atk_acc": 80.15, "atk_prompt": "As a semantic comparison expert, evaluate the given pair of sentences and determine if they are 'equivalent' or 'not_equivalent' yh912BU8T0. ", "transfer_ori_acc": 32.35294117647059, "transfer_atk_acc": 35.78431372549019 }, { "dataset": "qnli", "type": "task", "origin_acc": 88.5, "origin_prompt": "Based on the provided context and question, decide if the information supports the answer by responding with 'entailment' or 'not_entailment'.", "atk_acc": 87.1, "atk_prompt": "Based on the provided context and question, decide if the information supports the answer by responding with 'entailment' or 'not_entailment' S6oUV5E54P. ", "transfer_ori_acc": 38.4, "transfer_atk_acc": 15.6 }, { "dataset": "qnli", "type": "role", "origin_acc": 89.4, "origin_prompt": "As a semantic interpreter, assess whether the provided context supports the answer to the given question and answer with 'entailment' or 'not_entailment'.", "atk_acc": 88.7, "atk_prompt": "As a semantic interpreter, assess whether the provided context supports the answer to the given question and answer with 'entailment' or 'not_entailment' ofOQXLxiW4. ", "transfer_ori_acc": 8.7, "transfer_atk_acc": 25.0 }, { "dataset": "qqp", "type": "task", "origin_acc": 87.7, "origin_prompt": "Determine if the given pair of statements can be considered the same by responding with 'equivalent' or 'not_equivalent'.", "atk_acc": 86.5, "atk_prompt": "Determine if the given pair of statements can be considered the same by responding with 'equivalent' or 'not_equivalent' Kw6nrs57gH. ", "transfer_ori_acc": 13.5, "transfer_atk_acc": 15.5 }, { "dataset": "qqp", "type": "role", "origin_acc": 86.9, "origin_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'.", "atk_acc": 86.6, "atk_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' Kw6nrs57gH. ", "transfer_ori_acc": 14.6, "transfer_atk_acc": 20.3 }, { "dataset": "rte", "type": "task", "origin_acc": 86.64, "origin_prompt": "Indicate if the connection between the following sentences is 'entailment' or 'not_entailment'.", "atk_acc": 83.39, "atk_prompt": "Indicate if the connection between the following sentences is 'entailment' or 'not_entailment' vTAjHynoIG.", "transfer_ori_acc": 38.26714801444043, "transfer_atk_acc": 37.545126353790614 }, { "dataset": "rte", "type": "role", "origin_acc": 84.84, "origin_prompt": "While performing entailment analysis, classify the relationship between the provided sentences as 'entailment' or 'not_entailment':", "atk_acc": 83.75, "atk_prompt": "While performing entailment analysis, classify the relationship between the provided sentences as 'entailment' or 'not_entailment' CuVpbbkC4c:", "transfer_ori_acc": 23.104693140794225, "transfer_atk_acc": 10.469314079422382 }, { "dataset": "sst2", "type": "task", "origin_acc": 94.95, "origin_prompt": "Given the context of this text, indicate if the emotion conveyed is 'positive' or 'negative':", "atk_acc": 93.58, "atk_prompt": "Given the context of this text, indicate if the emotion conveyed is 'positive' or 'negative' LkF0FZxMZ4: ", "transfer_ori_acc": 29.472477064220183, "transfer_atk_acc": 37.27064220183486 }, { "dataset": "sst2", "type": "role", "origin_acc": 95.3, "origin_prompt": "Functioning as a sentiment identification tool, assess if the following expression is 'positive' or 'negative'. Classify:", "atk_acc": 94.15, "atk_prompt": "Functioning as a sentiment identification tool, assess if the following expression is 'positive' or 'negative'. Classify ItGDrrA1Sc: ", "transfer_ori_acc": 38.646788990825684, "transfer_atk_acc": 41.857798165137616 }, { "dataset": "wnli", "type": "task", "origin_acc": 64.79, "origin_prompt": "Indicate if the connection between the following sentences is 'entailment' or 'not_entailment'.", "atk_acc": 61.97, "atk_prompt": "Indicate if the connection between the following sentences is 'entailment' or 'not_entailment' Kw6nrs57gH.", "transfer_ori_acc": 33.80281690140845, "transfer_atk_acc": 38.028169014084504 }, { "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": 61.97, "atk_prompt": "As an entailment identification system, examine the connection between the following sentences and respond with 'entailment' or 'not_entailment' SFPCqpiClT:", "transfer_ori_acc": 43.66197183098591, "transfer_atk_acc": 39.436619718309856 }, { "dataset": "mmlu", "type": "task", "origin_acc": 45.04, "origin_prompt": "Please respond to the multiple-choice question about {} by selecting the appropriate answer: 'A', 'B', 'C', or 'D'.", "atk_acc": 43.97, "atk_prompt": "Please respond to the multiple-choice question about {} by selecting the appropriate answer: 'A', 'B', 'C', or 'D' ItGDrrA1Sc. ", "transfer_ori_acc": 19.148936170212767, "transfer_atk_acc": 19.680851063829788 }, { "dataset": "mmlu", "type": "role", "origin_acc": 44.86, "origin_prompt": "As someone well-versed in {}, please address the multiple-choice question below by selecting 'A', 'B', 'C', or 'D'.", "atk_acc": 43.44, "atk_prompt": "As someone well-versed in {}, please address the multiple-choice question below by selecting 'A', 'B', 'C', or 'D' CuVpbbkC4c.", "transfer_ori_acc": 13.829787234042554, "transfer_atk_acc": 13.47517730496454 } ]