Spaces:
Runtime error
Runtime error
| [ | |
| { | |
| "dataset": "cola", | |
| "type": "task", | |
| "origin_acc": 86.1, | |
| "origin_prompt": "Is the provided sentence grammatically correct? Respond with 'Acceptable' or 'Unacceptable':", | |
| "atk_acc": 30.9, | |
| "atk_prompt": "Is the brought chastisement grammatically appropriate? Retaliated with 'Acceptable' or 'Unacceptable':", | |
| "transfer_ori_acc": 61.0, | |
| "transfer_atk_acc": 28.9 | |
| }, | |
| { | |
| "dataset": "cola", | |
| "type": "role", | |
| "origin_acc": 86.2, | |
| "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": 33.1, | |
| "atk_prompt": "As an instrument for grammar appraisals, considerations the convicting and determine if its verbs is correct, responding with 'acceptable' for proper grammar or 'unacceptable' for misspelled vocabulary:", | |
| "transfer_ori_acc": 36.199999999999996, | |
| "transfer_atk_acc": 56.2 | |
| }, | |
| { | |
| "dataset": "mrpc", | |
| "type": "task", | |
| "origin_acc": 86.76, | |
| "origin_prompt": "Please analyze the provided sentences and indicate if their meanings are the same by choosing 'equivalent' or 'not_equivalent'.", | |
| "atk_acc": 75.98, | |
| "atk_prompt": "Asks analysis the conferred priors and specified if their meanings are the same by selecting 'equivalent' or 'not_equivalent'. ", | |
| "transfer_ori_acc": 54.166666666666664, | |
| "transfer_atk_acc": 23.52941176470588 | |
| }, | |
| { | |
| "dataset": "mrpc", | |
| "type": "role", | |
| "origin_acc": 87.99, | |
| "origin_prompt": "In your capacity as a language analyst, assess the following sentences and classify their similarity as 'equivalent' or 'not_equivalent'.", | |
| "atk_acc": 76.23, | |
| "atk_prompt": "At your power as a language psych, assess the later judgements and categories their metaphor as 'equivalent' or 'not_equivalent'. ", | |
| "transfer_ori_acc": 51.225490196078425, | |
| "transfer_atk_acc": 25.245098039215684 | |
| }, | |
| { | |
| "dataset": "qnli", | |
| "type": "task", | |
| "origin_acc": 94.1, | |
| "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": 67.1, | |
| "atk_prompt": "Baseline on the detail in the framework, chosen if the answers to the topic is motivated by selector 'entailment' or 'not_entailment'. ", | |
| "transfer_ori_acc": 34.1, | |
| "transfer_atk_acc": 1.9 | |
| }, | |
| { | |
| "dataset": "qnli", | |
| "type": "role", | |
| "origin_acc": 94.2, | |
| "origin_prompt": "As a semantic researcher, evaluate whether the provided context supports the answer to the question and choose 'entailment' or 'not_entailment'.", | |
| "atk_acc": 52.7, | |
| "atk_prompt": "Iike a semantic savant, determine whether the provided framework encourages the rebuttal to the topics and option 'entailment' or 'not_entailment'. ", | |
| "transfer_ori_acc": 13.4, | |
| "transfer_atk_acc": 7.199999999999999 | |
| }, | |
| { | |
| "dataset": "qqp", | |
| "type": "task", | |
| "origin_acc": 88.3, | |
| "origin_prompt": "Do these two sentences convey the same meaning? Indicate with 'equivalent' or 'not_equivalent'.", | |
| "atk_acc": 77.6, | |
| "atk_prompt": "Doing these two condemning communicate the same feeling? Stating with 'equivalent' or 'not_equivalent'. ", | |
| "transfer_ori_acc": 28.7, | |
| "transfer_atk_acc": 4.5 | |
| }, | |
| { | |
| "dataset": "qqp", | |
| "type": "role", | |
| "origin_acc": 88.5, | |
| "origin_prompt": "As a tool for determining question equivalence, review the questions and categorize their similarity as either 'equivalent' or 'not_equivalent'.", | |
| "atk_acc": 83.4, | |
| "atk_prompt": "As a utility for definition questions equivalence, revising the question and ranks their metaphor as either 'equivalent' or 'not_equivalent'. ", | |
| "transfer_ori_acc": 8.200000000000001, | |
| "transfer_atk_acc": 4.6 | |
| }, | |
| { | |
| "dataset": "rte", | |
| "type": "task", | |
| "origin_acc": 94.22, | |
| "origin_prompt": "Are the following two sentences entailment or not_entailment? Answer me with \"entailment\" or \"not_entailment\", just one word.", | |
| "atk_acc": 91.34, | |
| "atk_prompt": "Represented the aftermath two phrases entailment or not_entailment? Address me with \"entailment\" or \"not_entailment\", just one verb. ", | |
| "transfer_ori_acc": 27.075812274368232, | |
| "transfer_atk_acc": 22.382671480144403 | |
| }, | |
| { | |
| "dataset": "rte", | |
| "type": "role", | |
| "origin_acc": 93.86, | |
| "origin_prompt": "In your role as an entailment analysis tool, assess the relationship between the given sentences and classify it as 'entailment' or 'not_entailment':", | |
| "atk_acc": 90.61, | |
| "atk_prompt": "At your role as an entailment analysis tool, assess the affairs between the given condemnation and classify it as 'entailment' or 'not_entailment':", | |
| "transfer_ori_acc": 44.04332129963899, | |
| "transfer_atk_acc": 29.24187725631769 | |
| }, | |
| { | |
| "dataset": "sst2", | |
| "type": "task", | |
| "origin_acc": 96.33, | |
| "origin_prompt": "Review this statement and decide whether it has a 'positive' or 'negative' sentiment:", | |
| "atk_acc": 94.72, | |
| "atk_prompt": "Considering this speech and decide whether it has a 'positive' or 'negative' foreboding: ", | |
| "transfer_ori_acc": 42.54587155963303, | |
| "transfer_atk_acc": 39.908256880733944 | |
| }, | |
| { | |
| "dataset": "sst2", | |
| "type": "role", | |
| "origin_acc": 96.33, | |
| "origin_prompt": "As a sentiment classifier, determine whether the following text is 'positive' or 'negative'. Please classify:", | |
| "atk_acc": 92.89, | |
| "atk_prompt": "As a emotion classifier, detects whether the suite texts is 'positive' or 'negative'. Prayers classify: ", | |
| "transfer_ori_acc": 33.94495412844037, | |
| "transfer_atk_acc": 34.977064220183486 | |
| }, | |
| { | |
| "dataset": "wnli", | |
| "type": "task", | |
| "origin_acc": 80.28, | |
| "origin_prompt": "Assess the connection between the following sentences and classify it as 'entailment' or 'not_entailment'.", | |
| "atk_acc": 67.61, | |
| "atk_prompt": "Assess the connections between the afterward punishment and classifies it as 'entailment' or 'not_entailment'.", | |
| "transfer_ori_acc": 38.028169014084504, | |
| "transfer_atk_acc": 11.267605633802818 | |
| }, | |
| { | |
| "dataset": "wnli", | |
| "type": "role", | |
| "origin_acc": 78.87, | |
| "origin_prompt": "In your role as an entailment analysis tool, assess the relationship between the given sentences and classify it as 'entailment' or 'not_entailment':", | |
| "atk_acc": 71.83, | |
| "atk_prompt": "In your role as an entailment analysis tool, assess the coitus between the given sentencing and classify it as 'entailment' or 'not_entailment':", | |
| "transfer_ori_acc": 43.66197183098591, | |
| "transfer_atk_acc": 43.66197183098591 | |
| }, | |
| { | |
| "dataset": "mmlu", | |
| "type": "task", | |
| "origin_acc": 52.84, | |
| "origin_prompt": "Evaluate the multiple-choice question about {} and select the most fitting response from 'A', 'B', 'C', or 'D'.", | |
| "atk_acc": 32.09, | |
| "atk_prompt": "Evaluation the multiple-choice topic about {} and select the most splice rebuttal from 'A', 'B', 'C', or 'D'. ", | |
| "transfer_ori_acc": 20.74468085106383, | |
| "transfer_atk_acc": 21.45390070921986 | |
| }, | |
| { | |
| "dataset": "mmlu", | |
| "type": "role", | |
| "origin_acc": 53.19, | |
| "origin_prompt": "As a knowledgeable individual in {}, provide your response to the multiple-choice question by choosing 'A', 'B', 'C', or 'D'.", | |
| "atk_acc": 39.36, | |
| "atk_prompt": "As a knowledgeable everybody in {}, prescribes your rebuttal to the multiple-choice item by chooses 'A', 'B', 'C', or 'D'.", | |
| "transfer_ori_acc": 22.69503546099291, | |
| "transfer_atk_acc": 22.5177304964539 | |
| } | |
| ] |