Spaces:
Runtime error
Runtime error
| [ | |
| { | |
| "dataset": "cola", | |
| "type": "task", | |
| "origin_acc": 86.4, | |
| "origin_prompt": "Review the sentence below and identify whether its grammar is 'Acceptable' or 'Unacceptable'.", | |
| "atk_acc": 85.6, | |
| "atk_prompt": "Review the sentence below and identify whether its grammar is 'Acceptable' or 'Unacceptable' Q47Jmd4lMV. ", | |
| "transfer_ori_acc": 26.0, | |
| "transfer_atk_acc": 10.6 | |
| }, | |
| { | |
| "dataset": "cola", | |
| "type": "role", | |
| "origin_acc": 86.6, | |
| "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": 86.2, | |
| "atk_prompt": "While performing grammar analysis, classify the grammar of the following sentence as 'acceptable' for correct grammar or 'unacceptable' for incorrect grammar EAB4KP2NVY. ", | |
| "transfer_ori_acc": 49.9, | |
| "transfer_atk_acc": 61.3 | |
| }, | |
| { | |
| "dataset": "mrpc", | |
| "type": "task", | |
| "origin_acc": 86.27, | |
| "origin_prompt": "Do these two sentences have the same underlying meaning? Respond with 'equivalent' or 'not_equivalent'.", | |
| "atk_acc": 85.78, | |
| "atk_prompt": "Do these two sentences have the same underlying meaning? Respond with 'equivalent' or 'not_equivalent' 6a4Yn3RGVc. ", | |
| "transfer_ori_acc": 60.5392156862745, | |
| "transfer_atk_acc": 66.66666666666666 | |
| }, | |
| { | |
| "dataset": "mrpc", | |
| "type": "role", | |
| "origin_acc": 87.5, | |
| "origin_prompt": "As a textual similarity evaluator, analyze the following pair of sentences and determine if they are semantically 'equivalent' or 'not_equivalent'.", | |
| "atk_acc": 87.25, | |
| "atk_prompt": "As a textual similarity evaluator, analyze the following pair of sentences and determine if they are semantically 'equivalent' or 'not_equivalent' ofw9fEkN5R. ", | |
| "transfer_ori_acc": 59.31372549019608, | |
| "transfer_atk_acc": 65.68627450980392 | |
| }, | |
| { | |
| "dataset": "qnli", | |
| "type": "task", | |
| "origin_acc": 94.0, | |
| "origin_prompt": "Analyze the context and question, and indicate if the context entails the answer by choosing 'entailment' or 'not_entailment'.", | |
| "atk_acc": 93.4, | |
| "atk_prompt": "Analyze the context and question, and indicate if the context entails the answer by choosing 'entailment' or 'not_entailment' lf8wBa2yEm. ", | |
| "transfer_ori_acc": 35.0, | |
| "transfer_atk_acc": 16.1 | |
| }, | |
| { | |
| "dataset": "qnli", | |
| "type": "role", | |
| "origin_acc": 93.7, | |
| "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": 93.5, | |
| "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' zq0DcZ5dnI. ", | |
| "transfer_ori_acc": 24.6, | |
| "transfer_atk_acc": 1.2 | |
| }, | |
| { | |
| "dataset": "qqp", | |
| "type": "task", | |
| "origin_acc": 88.5, | |
| "origin_prompt": "Are the following two questions equivalent or not? Answer me with \"equivalent\" or \"not_equivalent\".", | |
| "atk_acc": 88.4, | |
| "atk_prompt": "Are the following two questions equivalent or not? Answer me with \"equivalent\" or \"not_equivalent Cev0s49fbe\". ", | |
| "transfer_ori_acc": 34.8, | |
| "transfer_atk_acc": 32.1 | |
| }, | |
| { | |
| "dataset": "qqp", | |
| "type": "role", | |
| "origin_acc": 89.1, | |
| "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": 88.7, | |
| "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' KjPJJ2a7RB. ", | |
| "transfer_ori_acc": 32.9, | |
| "transfer_atk_acc": 28.599999999999998 | |
| }, | |
| { | |
| "dataset": "rte", | |
| "type": "task", | |
| "origin_acc": 93.5, | |
| "origin_prompt": "Identify whether the given pair of sentences demonstrates entailment or not_entailment. Answer with 'entailment' or 'not_entailment'.", | |
| "atk_acc": 92.78, | |
| "atk_prompt": "Identify whether the given pair of sentences demonstrates entailment or not_entailment. Answer with 'entailment' or 'not_entailment' d6ZQ3u0GBQ. ", | |
| "transfer_ori_acc": 7.2202166064981945, | |
| "transfer_atk_acc": 5.054151624548736 | |
| }, | |
| { | |
| "dataset": "rte", | |
| "type": "role", | |
| "origin_acc": 93.5, | |
| "origin_prompt": "Acting as an entailment detection instrument, determine if the given pair of sentences demonstrates entailment or not_entailment. Answer with 'entailment' or 'not_entailment'.", | |
| "atk_acc": 93.14, | |
| "atk_prompt": "Acting as an entailment detection instrument, determine if the given pair of sentences demonstrates entailment or not_entailment. Answer with 'entailment' or 'not_entailment' BWmsLx8xOA. ", | |
| "transfer_ori_acc": 4.332129963898916, | |
| "transfer_atk_acc": 46.20938628158845 | |
| }, | |
| { | |
| "dataset": "sst2", | |
| "type": "task", | |
| "origin_acc": 96.79, | |
| "origin_prompt": "Evaluate the sentiment of the given text and classify it as 'positive' or 'negative'.", | |
| "atk_acc": 96.56, | |
| "atk_prompt": "Evaluate the sentiment of the given text and classify it as 'positive' or 'negative' Cev0s49fbe. ", | |
| "transfer_ori_acc": 10.435779816513762, | |
| "transfer_atk_acc": 9.63302752293578 | |
| }, | |
| { | |
| "dataset": "sst2", | |
| "type": "role", | |
| "origin_acc": 96.56, | |
| "origin_prompt": "As a sentiment classifier, determine whether the following text is 'positive' or 'negative'.", | |
| "atk_acc": 96.44, | |
| "atk_prompt": "As a sentiment classifier, determine whether the following text is 'positive' or 'negative' 6a4Yn3RGVc. ", | |
| "transfer_ori_acc": 15.36697247706422, | |
| "transfer_atk_acc": 14.449541284403669 | |
| }, | |
| { | |
| "dataset": "wnli", | |
| "type": "task", | |
| "origin_acc": 77.46, | |
| "origin_prompt": "Assess the connection between the following sentences and classify it as 'entailment' or 'not_entailment'.", | |
| "atk_acc": 76.06, | |
| "atk_prompt": "Assess the connection between the following sentences and classify it as 'entailment' or 'not_entailment' 5yWbBXztUY. ", | |
| "transfer_ori_acc": 19.718309859154928, | |
| "transfer_atk_acc": 18.30985915492958 | |
| }, | |
| { | |
| "dataset": "wnli", | |
| "type": "role", | |
| "origin_acc": 76.06, | |
| "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": 76.06, | |
| "atk_prompt": "In your role as an entailment analysis tool, assess the relationship between the given sentences and classify it as 'entailment' or 'not_entailment' 5XeflW1ZJc. ", | |
| "transfer_ori_acc": 23.943661971830984, | |
| "transfer_atk_acc": 12.676056338028168 | |
| }, | |
| { | |
| "dataset": "mmlu", | |
| "type": "task", | |
| "origin_acc": 53.9, | |
| "origin_prompt": "Below are multiple-choice question concerning {}. Indicate your response with 'A', 'B', 'C', or 'D'.", | |
| "atk_acc": 52.84, | |
| "atk_prompt": "Below are multiple-choice question concerning {}. Indicate your response with 'A', 'B', 'C', or 'D' RcRneWo6Iv. ", | |
| "transfer_ori_acc": 6.382978723404255, | |
| "transfer_atk_acc": 6.914893617021277 | |
| }, | |
| { | |
| "dataset": "mmlu", | |
| "type": "role", | |
| "origin_acc": 54.08, | |
| "origin_prompt": "Considering your familiarity with {}, attend to the following multiple-choice question by picking 'A', 'B', 'C', or 'D'.", | |
| "atk_acc": 53.01, | |
| "atk_prompt": "Considering your familiarity with {}, attend to the following multiple-choice question by picking 'A', 'B', 'C', or 'D' Q47Jmd4lMV.", | |
| "transfer_ori_acc": 8.156028368794328, | |
| "transfer_atk_acc": 6.914893617021277 | |
| } | |
| ] |