results_transfer/google-flan-t5-large_google-flan-ul2/checklist_0_shot.json

[
    {
        "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": 86.1,
        "transfer_atk_acc": 84.0
    },
    {
        "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": 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": 86.3,
        "transfer_atk_acc": 83.89999999999999
    },
    {
        "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": 86.02941176470588,
        "transfer_atk_acc": 86.27450980392157
    },
    {
        "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": 84.31372549019608,
        "transfer_atk_acc": 84.80392156862744
    },
    {
        "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": 94.8,
        "transfer_atk_acc": 94.69999999999999
    },
    {
        "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": 94.6,
        "transfer_atk_acc": 94.5
    },
    {
        "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": 88.1,
        "transfer_atk_acc": 89.2
    },
    {
        "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": 88.0,
        "transfer_atk_acc": 88.9
    },
    {
        "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": 92.7797833935018,
        "transfer_atk_acc": 92.4187725631769
    },
    {
        "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": 93.50180505415162,
        "transfer_atk_acc": 92.4187725631769
    },
    {
        "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": 95.41284403669725,
        "transfer_atk_acc": 94.4954128440367
    },
    {
        "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": 95.9862385321101,
        "transfer_atk_acc": 95.87155963302753
    },
    {
        "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": 78.87323943661971,
        "transfer_atk_acc": 74.64788732394366
    },
    {
        "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": 80.28169014084507,
        "transfer_atk_acc": 76.05633802816901
    },
    {
        "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": 53.36879432624113,
        "transfer_atk_acc": 52.4822695035461
    },
    {
        "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": 52.659574468085104,
        "transfer_atk_acc": 53.01418439716312
    },
    {
        "dataset": "squad_v2",
        "type": "task",
        "origin_acc": 87.65,
        "origin_prompt": "Please derive the most fitting answer from the context. If there isn't an answer in the context, respond with 'unanswerable'.",
        "atk_acc": 86.82,
        "atk_prompt": "Please derive the most fitting answer from the context. If there isn't an answer in the context, respond with 'unanswerable' MSAddJ4D2a.",
        "transfer_ori_acc": 89.88229949874687,
        "transfer_atk_acc": 89.54896616541353
    },
    {
        "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": 87.15,
        "atk_prompt": "As an authority on the context, locate the most accurate answer. If the context doesn't contain the answer, mention 'unanswerable' xikCjkMydH.",
        "transfer_ori_acc": 88.98646616541353,
        "transfer_atk_acc": 87.58646616541355
    },
    {
        "dataset": "iwslt",
        "type": "task",
        "origin_acc": 0.14,
        "origin_prompt": "Please perform a translation of the given sentence, converting it from {} to {}:",
        "atk_acc": 0.1,
        "atk_prompt": "Please perform a translation of the given sentence, converting it from {} to mzjVogkRhM {}:",
        "transfer_ori_acc": 0.17287994792239936,
        "transfer_atk_acc": 0.15893556619118035
    },
    {
        "dataset": "iwslt",
        "type": "role",
        "origin_acc": 0.14,
        "origin_prompt": "Taking on the role of a certified translator, perform a translation of the given sentence from {} to {}:",
        "atk_acc": 0.08,
        "atk_prompt": "Taking on the role of a certified translator, perform a translation of the given sentence from {} to zexKUTWJYG {}: ",
        "transfer_ori_acc": 0.1727120569025014,
        "transfer_atk_acc": 0.1633255034209739
    },
    {
        "dataset": "un_multi",
        "type": "task",
        "origin_acc": 0.27,
        "origin_prompt": "Accurately translate the sentence from {} to {}, ensuring the meaning remains intact:",
        "atk_acc": 0.21,
        "atk_prompt": "Accurately translate the sentence from {} to {}, ensuring the meaning remains intact EAB4KP2NVY:",
        "transfer_ori_acc": 0.31735504988957897,
        "transfer_atk_acc": 0.3098482661555996
    },
    {
        "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": "Assuming the role of an expert language translator, translate the given text from {} to vTAjHynoIG {}: ",
        "transfer_ori_acc": 0.31373650381115664,
        "transfer_atk_acc": 0.3103473904869731
    },
    {
        "dataset": "math",
        "type": "task",
        "origin_acc": 16.25,
        "origin_prompt": "Evaluate the following mathematical problem that includes {}:",
        "atk_acc": 10.62,
        "atk_prompt": "Evaluate the following mathematical problem that includes is77sOXAu8 {}:",
        "transfer_ori_acc": 13.750000000000002,
        "transfer_atk_acc": 12.5
    },
    {
        "dataset": "math",
        "type": "role",
        "origin_acc": 14.37,
        "origin_prompt": "As a math guru, evaluate the following mathematical task regarding {}:",
        "atk_acc": 10.0,
        "atk_prompt": "As a math guru, evaluate the following mathematical task regarding KjPJJ2a7RB {}: ",
        "transfer_ori_acc": 15.0,
        "transfer_atk_acc": 13.125
    }
]