[ { "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": 32.1, "atk_prompt": "Examining the given condemnation and decide if it is pronoun sounds. Answer with neither 'Acceptable' or 'Unacceptable':", "transfer_ori_acc": 50.0, "transfer_atk_acc": 32.4 }, { "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": 46.8, "atk_prompt": "Because an devices for grammar assessment, consdier the punishments and determining if its grammar is correct, responԁing with 'acceptable' for correct grammar or 'unacceptable' for incorrect pronunciation:", "transfer_ori_acc": 36.199999999999996, "transfer_atk_acc": 50.8 }, { "dataset": "mrpc", "type": "task", "origin_acc": 82.11, "origin_prompt": "Do the meanings of these two statements align? Indicate your answer with 'equivalent' or 'not_equivalent'.", "atk_acc": 75.25, "atk_prompt": "Do the meanings of these two statemеnts alіgn? Denotes your answer with 'equivalent' or 'not_equivalent'. ", "transfer_ori_acc": 45.34313725490196, "transfer_atk_acc": 40.931372549019606 }, { "dataset": "mrpc", "type": "role", "origin_acc": 81.62, "origin_prompt": "As a language comparison expert, examine the given pair of sentences and decide if their meanings align, answering with 'equivalent' or 'not_equivalent'.", "atk_acc": 74.26, "atk_prompt": "As a language comparison e×pert, examine the given p air of sentences and decide if their meaings alin, answering with 'equivalent' or 'not_equivalent'. ", "transfer_ori_acc": 63.23529411764706, "transfer_atk_acc": 23.52941176470588 }, { "dataset": "qnli", "type": "task", "origin_acc": 89.8, "origin_prompt": "Evaluate whether the given context supports the answer to the question by responding with 'entailment' or 'not_entailment'.", "atk_acc": 56.8, "atk_prompt": "Evɑluate lf the given context supports the replying to the question by answer with 'entailment' or 'not_entailment'. ", "transfer_ori_acc": 26.5, "transfer_atk_acc": 12.6 }, { "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": 50.8, "atk_prompt": "Since a semantic interpreter, as sess lf the provіded background supports the answer to the ɡiven question and answer with 'entailment' or 'not_entailment'. ", "transfer_ori_acc": 8.7, "transfer_atk_acc": 7.3999999999999995 }, { "dataset": "qqp", "type": "task", "origin_acc": 88.1, "origin_prompt": "Are the following two questions equivalent or not? Answer me with \"equivalent\" or \"not_equivalent\".", "atk_acc": 78.6, "atk_prompt": "Are the following tԝo matters equivalent or not? Replying me with \"equivalent\" or \"not_equivalent\". ", "transfer_ori_acc": 23.9, "transfer_atk_acc": 18.5 }, { "dataset": "qqp", "type": "role", "origin_acc": 87.7, "origin_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.", "atk_acc": 82.5, "atk_prompt": "Actinɡ as a question equivaleոce inst rument, determine if the prvided questions are equivalent in mean, answering with 'equivalent' for similar questions or 'not_equivalent' for dіssimilar ones. ", "transfer_ori_acc": 25.0, "transfer_atk_acc": 6.0 }, { "dataset": "rte", "type": "task", "origin_acc": 86.64, "origin_prompt": "Are the following two sentences entailment or not_entailment? Answer me with \"entailment\" or \"not_entailment\", just one word.", "atk_acc": 82.31, "atk_prompt": "Arе the following to sentences entailment or not_entailment? Respond me with \"entailment\" or \"not_entailment\", just one word. ", "transfer_ori_acc": 27.075812274368232, "transfer_atk_acc": 7.9422382671480145 }, { "dataset": "rte", "type": "role", "origin_acc": 85.56, "origin_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':", "atk_acc": 82.67, "atk_prompt": "As an instrument for entailment evaluation, consider the t wo sentences and determine if their relationѕhip is 'entailment' or 'not_entailment'. Respond with 'entailment' or 'not_entailment':", "transfer_ori_acc": 24.90974729241877, "transfer_atk_acc": 24.187725631768952 }, { "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": 76.26, "atk_prompt": "Giveո the background of this text, depicts if the emotions transmitted is 'positive' or 'negative': ", "transfer_ori_acc": 29.472477064220183, "transfer_atk_acc": 42.6605504587156 }, { "dataset": "sst2", "type": "role", "origin_acc": 95.3, "origin_prompt": "Serving as a sentiment evaluation model, determine if the given statement is 'positive' or 'negative'. Classify:", "atk_acc": 88.53, "atk_prompt": "Serving as a emotion evluation models, determining if the conferred state ment is 'positive' or 'negative'. Classify: ", "transfer_ori_acc": 26.261467889908257, "transfer_atk_acc": 21.3302752293578 }, { "dataset": "wnli", "type": "task", "origin_acc": 66.2, "origin_prompt": "Analyze the two provided sentences and decide if their relationship is 'entailment' or 'not_entailment'.", "atk_acc": 57.75, "atk_prompt": "Analyse the two provided sentences and dеcide if their relations is 'entailment' or 'not_entailment'.", "transfer_ori_acc": 7.042253521126761, "transfer_atk_acc": 7.042253521126761 }, { "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": 57.75, "atk_prompt": "As an entailment identification system, investigated the connection between the subsequently sentences and replying with 'entailment' or 'not_entailment':", "transfer_ori_acc": 43.66197183098591, "transfer_atk_acc": 39.436619718309856 }, { "dataset": "mmlu", "type": "task", "origin_acc": 45.74, "origin_prompt": "For each multiple-choice question about {}, identify the correct answer by selecting 'A', 'B', 'C', or 'D'.", "atk_acc": 43.26, "atk_prompt": "For each multip le-choice question about {}, identify the rectify answer by selеcting 'A', 'B', 'C', or 'D'. ", "transfer_ori_acc": 22.69503546099291, "transfer_atk_acc": 25.354609929078016 }, { "dataset": "mmlu", "type": "role", "origin_acc": 44.86, "origin_prompt": "Utilizing your expertise in {}, answer the following multiple-choice question by picking 'A', 'B', 'C', or 'D'.", "atk_acc": 40.96, "atk_prompt": "Utilizіng your expertise in {}, answeⲅ the following multiple-choice question by collection 'A', 'B', 'C', or 'D'.", "transfer_ori_acc": 20.74468085106383, "transfer_atk_acc": 33.51063829787234 } ]