Spaces:
Runtime error
Runtime error
| [ | |
| { | |
| "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": 86.1, | |
| "transfer_atk_acc": 78.0 | |
| }, | |
| { | |
| "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": 86.3, | |
| "transfer_atk_acc": 82.69999999999999 | |
| }, | |
| { | |
| "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": 85.5392156862745, | |
| "transfer_atk_acc": 85.7843137254902 | |
| }, | |
| { | |
| "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": 86.27450980392157, | |
| "transfer_atk_acc": 87.00980392156863 | |
| }, | |
| { | |
| "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": 93.89999999999999, | |
| "transfer_atk_acc": 93.4 | |
| }, | |
| { | |
| "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": 94.6, | |
| "transfer_atk_acc": 93.0 | |
| }, | |
| { | |
| "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": 87.8, | |
| "transfer_atk_acc": 88.1 | |
| }, | |
| { | |
| "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": 87.9, | |
| "transfer_atk_acc": 87.7 | |
| }, | |
| { | |
| "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": 94.22382671480143, | |
| "transfer_atk_acc": 93.86281588447653 | |
| }, | |
| { | |
| "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": 93.50180505415162, | |
| "transfer_atk_acc": 92.7797833935018 | |
| }, | |
| { | |
| "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": 95.41284403669725, | |
| "transfer_atk_acc": 94.95412844036697 | |
| }, | |
| { | |
| "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": 96.21559633027523, | |
| "transfer_atk_acc": 94.95412844036697 | |
| }, | |
| { | |
| "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": 77.46478873239437, | |
| "transfer_atk_acc": 77.46478873239437 | |
| }, | |
| { | |
| "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": 80.28169014084507, | |
| "transfer_atk_acc": 78.87323943661971 | |
| }, | |
| { | |
| "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": 51.77304964539007, | |
| "transfer_atk_acc": 53.01418439716312 | |
| }, | |
| { | |
| "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": 52.12765957446809, | |
| "transfer_atk_acc": 53.72340425531915 | |
| }, | |
| { | |
| "dataset": "squad_v2", | |
| "type": "task", | |
| "origin_acc": 87.55, | |
| "origin_prompt": "Find the correct answer in the context provided. If an answer cannot be found, please respond with 'unanswerable'.", | |
| "atk_acc": 85.82, | |
| "atk_prompt": "Find the corrected replying in the ϲontext given. If an answer cannot be found, please respond with 'unanswerable'.", | |
| "transfer_ori_acc": 89.98646616541353, | |
| "transfer_atk_acc": 90.54896616541353 | |
| }, | |
| { | |
| "dataset": "squad_v2", | |
| "type": "role", | |
| "origin_acc": 88.05, | |
| "origin_prompt": "As an authority on the context, locate the most accurate answer. If the context doesn't contain the answer, mention 'unanswerable'.", | |
| "atk_acc": 81.52, | |
| "atk_prompt": "As an authoⲅity on the context, locate the most meticulous respond. Whether the context doesn't contain the answer, mentioո 'unanswerable'.", | |
| "transfer_ori_acc": 88.98646616541353, | |
| "transfer_atk_acc": 88.70729949874686 | |
| }, | |
| { | |
| "dataset": "iwslt", | |
| "type": "task", | |
| "origin_acc": 0.14, | |
| "origin_prompt": "Please perform a translation of the given sentence, converting it from {} to {}:", | |
| "atk_acc": 0.13, | |
| "atk_prompt": "Please perfоrm a trɑnslation of the given sentence, converting it from {} to {}:", | |
| "transfer_ori_acc": 0.17287994792239936, | |
| "transfer_atk_acc": 0.17384832122858687 | |
| }, | |
| { | |
| "dataset": "iwslt", | |
| "type": "role", | |
| "origin_acc": 0.14, | |
| "origin_prompt": "In the capacity of a proficient translator, please change the following sentence from {} to {}:", | |
| "atk_acc": 0.12, | |
| "atk_prompt": "In the capacities of a proficient traոslator, рlease change the following sentence from {} to {}: ", | |
| "transfer_ori_acc": 0.17223574608131062, | |
| "transfer_atk_acc": 0.16503387004479023 | |
| }, | |
| { | |
| "dataset": "un_multi", | |
| "type": "task", | |
| "origin_acc": 0.27, | |
| "origin_prompt": "Accurately translate the sentence from {} to {}, ensuring the meaning remains intact:", | |
| "atk_acc": 0.17, | |
| "atk_prompt": "Scrupulously translate the sentence from {} to {}, ensuring the meani ng remains intact:", | |
| "transfer_ori_acc": 0.31735504988957897, | |
| "transfer_atk_acc": 0.3150453261764125 | |
| }, | |
| { | |
| "dataset": "un_multi", | |
| "type": "role", | |
| "origin_acc": 0.27, | |
| "origin_prompt": "Assuming the role of an expert language translator, translate the given text from {} to {}:", | |
| "atk_acc": 0.19, | |
| "atk_prompt": "Asѕuming the role of an experts la nguage translator, translate the ɡiven text from {} to {}: ", | |
| "transfer_ori_acc": 0.31373650381115664, | |
| "transfer_atk_acc": 0.2448382580575487 | |
| }, | |
| { | |
| "dataset": "math", | |
| "type": "task", | |
| "origin_acc": 16.25, | |
| "origin_prompt": "Evaluate the following mathematical problem that includes {}:", | |
| "atk_acc": 10.0, | |
| "atk_prompt": "Evaluate the following mathematical problem that inϲludes {}:", | |
| "transfer_ori_acc": 13.750000000000002, | |
| "transfer_atk_acc": 14.374999999999998 | |
| }, | |
| { | |
| "dataset": "math", | |
| "type": "role", | |
| "origin_acc": 13.12, | |
| "origin_prompt": "As a mathematics instructor, calculate the answer to the following problem related to {}:", | |
| "atk_acc": 9.38, | |
| "atk_prompt": "As a mathematics instructoⲅ, calculate the answers to the following problem related to {}: ", | |
| "transfer_ori_acc": 15.0, | |
| "transfer_atk_acc": 13.750000000000002 | |
| } | |
| ] |