results_transfer/google-flan-ul2_chatgpt/textbugger_3_shot.json

[
    {
        "dataset": "cola",
        "type": "task",
        "origin_acc": 86.4,
        "origin_prompt": "Examine the given sentence and decide if it is grammatically sound. Answer with either 'Acceptable' or 'Unacceptable'.",
        "atk_acc": 68.1,
        "atk_prompt": "Examine the given condemnation and decide if it is pronouns sound. Answer with either 'Acceptable' or 'Unacceptable'. ",
        "transfer_ori_acc": 81.0,
        "transfer_atk_acc": 78.5
    },
    {
        "dataset": "cola",
        "type": "role",
        "origin_acc": 86.9,
        "origin_prompt": "In the capacity of a grammar assessment system, indicate if the structure of the provided sentence is grammatically correct, responding with 'acceptable' or 'unacceptable'.",
        "atk_acc": 85.0,
        "atk_prompt": "In the ability of a grammar assessment system, denotes if the structure of the provided sentence is grammatical corrected, responding with 'acceptable' or 'unacceptable'. ",
        "transfer_ori_acc": 80.5,
        "transfer_atk_acc": 79.5
    },
    {
        "dataset": "mrpc",
        "type": "task",
        "origin_acc": 87.01,
        "origin_prompt": "Can the given sentences be considered semantically identical? Please reply with 'equivalent' or 'not_equivalent'.",
        "atk_acc": 83.82,
        "atk_prompt": "Can the given sentences be deemed semantically idntical? Please reply with 'equivalent' or 'not_equivalent'. ",
        "transfer_ori_acc": 72.0,
        "transfer_atk_acc": 72.0
    },
    {
        "dataset": "mrpc",
        "type": "role",
        "origin_acc": 87.5,
        "origin_prompt": "In your capacity as a language analyst, assess the following sentences and classify their similarity as 'equivalent' or 'not_equivalent'.",
        "atk_acc": 83.58,
        "atk_prompt": "In your capacity as a language ɑnalyst, assess the following sentenes and classify their metaphor as 'equivalent' or 'not_equivalent'. ",
        "transfer_ori_acc": 62.5,
        "transfer_atk_acc": 64.0
    },
    {
        "dataset": "qnli",
        "type": "task",
        "origin_acc": 94.0,
        "origin_prompt": "Review the given context and question, and decide if the context contains enough information to support the answer by selecting 'entailment' or 'not_entailment'.",
        "atk_acc": 92.9,
        "atk_prompt": "Review the given coոtext and question, and dcide if the context contains enough particulars to support the answeⲅ by selecting 'entailment' or 'not_entailment'. ",
        "transfer_ori_acc": 83.0,
        "transfer_atk_acc": 84.0
    },
    {
        "dataset": "qnli",
        "type": "role",
        "origin_acc": 93.7,
        "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": 92.3,
        "atk_prompt": "Iike a semantic interpretation, evaluated lf the prоvided context supports the answer to the given question and answers with 'entailment' or 'not_entailment'. ",
        "transfer_ori_acc": 83.5,
        "transfer_atk_acc": 81.5
    },
    {
        "dataset": "qqp",
        "type": "task",
        "origin_acc": 88.8,
        "origin_prompt": "Determine if the given pair of statements can be considered the same by responding with 'equivalent' or 'not_equivalent'.",
        "atk_acc": 87.6,
        "atk_prompt": "Defining if the given paired of statements can be considered the same by responds with 'equivalent' or 'not_equivalent'. ",
        "transfer_ori_acc": 82.5,
        "transfer_atk_acc": 83.0
    },
    {
        "dataset": "qqp",
        "type": "role",
        "origin_acc": 90.0,
        "origin_prompt": "As a tool for determining question equivalence, review the questions and categorize their similarity as either 'equivalent' or 'not_equivalent'.",
        "atk_acc": 87.6,
        "atk_prompt": "As a tool for deciding queѕtion equivalence, review the questions and classifications their portrait as neither 'equivalent' or 'not_equivalent'. ",
        "transfer_ori_acc": 83.5,
        "transfer_atk_acc": 83.0
    },
    {
        "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": 92.42,
        "atk_prompt": "Identifying whether the given pair of sentencing demonstrates entailment or not_entailment. Respond with 'entailment' or 'not_entailment'. ",
        "transfer_ori_acc": 82.0,
        "transfer_atk_acc": 82.5
    },
    {
        "dataset": "rte",
        "type": "role",
        "origin_acc": 93.86,
        "origin_prompt": "Working as an entailment classifier, identify whether the given pair of sentences displays entailment or not_entailment. Respond with 'entailment' or 'not_entailment'.",
        "atk_acc": 90.97,
        "atk_prompt": "Working as an entailment classifier, identify both the given piar of sentences displaying entailment or not_entailment. Respond with 'entailment' or 'not_entailment'. ",
        "transfer_ori_acc": 81.0,
        "transfer_atk_acc": 80.5
    },
    {
        "dataset": "sst2",
        "type": "task",
        "origin_acc": 96.79,
        "origin_prompt": "Determine the overall sentiment of this sentence, categorizing it as 'positive' or 'negative'.",
        "atk_acc": 95.87,
        "atk_prompt": "Determine the general sentiment of this condemnation, categorizing it as 'positive' or 'negative'. ",
        "transfer_ori_acc": 95.0,
        "transfer_atk_acc": 96.5
    },
    {
        "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": 95.18,
        "atk_prompt": "As a emotion classifier, determine whether the follоwing wording is 'positive' or 'negative'. ",
        "transfer_ori_acc": 97.0,
        "transfer_atk_acc": 96.0
    },
    {
        "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": "Assss the connection between the following sentences and classify it as 'entailment' or 'not_entailment'. ",
        "transfer_ori_acc": 63.38028169014085,
        "transfer_atk_acc": 64.7887323943662
    },
    {
        "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": 71.83,
        "atk_prompt": "In your role as an entailment analysis tool, assess the relationship between the given punishments and categorize it as 'entailment' or 'not_entailment'. ",
        "transfer_ori_acc": 60.56338028169014,
        "transfer_atk_acc": 64.7887323943662
    },
    {
        "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": 52.3,
        "atk_prompt": "Belоw are multipl-echoice quеstion concening {}. Indicate your response with 'A', 'B', 'C', or 'D'. ",
        "transfer_ori_acc": 66.0,
        "transfer_atk_acc": 66.0
    },
    {
        "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": 52.48,
        "atk_prompt": "Considering your familiarity with {}, attend to the following mutliple-choice matter by picking 'A', 'B', 'C', or 'D'.",
        "transfer_ori_acc": 65.5,
        "transfer_atk_acc": 65.5
    },
    {
        "dataset": "squad_v2",
        "type": "task",
        "origin_acc": 91.05,
        "origin_prompt": "Identify the most relevant answer from the context. If it's not possible to find an answer, state 'unanswerable'.",
        "atk_acc": 89.38,
        "atk_prompt": "Identify the most reevant answer from the context. If it's not achievable to find an answer, state 'unanswerable'.",
        "transfer_ori_acc": 72.4741019997885,
        "transfer_atk_acc": 72.06799144367795
    },
    {
        "dataset": "squad_v2",
        "type": "role",
        "origin_acc": 90.55,
        "origin_prompt": "Being well-versed in the context, please derive the most fitting answer. If there isn't an answer in the context, use 'unanswerable'.",
        "atk_acc": 80.92,
        "atk_prompt": "Underway well-versed in the context, please subtract the most fitting answer. If there isn't an answer in the framework, use 'unanswerable'.",
        "transfer_ori_acc": 73.10444801603916,
        "transfer_atk_acc": 74.07004494573145
    },
    {
        "dataset": "iwslt",
        "type": "task",
        "origin_acc": 0.27,
        "origin_prompt": "Accurately translate the sentence from {} to {}, ensuring the meaning remains intact.",
        "atk_acc": 0.23,
        "atk_prompt": "Acϲurately translate the sentence from {} to {}, ensuring the meaning remains unaffected. ",
        "transfer_ori_acc": 0.2368663340254173,
        "transfer_atk_acc": 0.23523551258635614
    },
    {
        "dataset": "iwslt",
        "type": "role",
        "origin_acc": 0.27,
        "origin_prompt": "Taking on the role of a certified translator, perform a translation of the given sentence from {} to {}.",
        "atk_acc": 0.23,
        "atk_prompt": "Taking on the role of a certifying translators, pefrorm a translatiоn of the givеn sentence from {} to {}. ",
        "transfer_ori_acc": 0.24270720862925335,
        "transfer_atk_acc": 0.24775340113684757
    },
    {
        "dataset": "un_multi",
        "type": "task",
        "origin_acc": 0.36,
        "origin_prompt": "Perform a machine translation for the provided sentence, changing it from {} to {}.",
        "atk_acc": 0.35,
        "atk_prompt": "Preform a machinery translation for the providеd sentence, changing it from {} to {}. ",
        "transfer_ori_acc": 0.3513571788256229,
        "transfer_atk_acc": 0.34794341664445116
    },
    {
        "dataset": "un_multi",
        "type": "role",
        "origin_acc": 0.36,
        "origin_prompt": "As a skilled linguist, please convert the following sentence from {} to {}.",
        "atk_acc": 0.35,
        "atk_prompt": "As a skillful linguist, pelase convеrt the following sentence from {} to {}. ",
        "transfer_ori_acc": 0.3505914536760883,
        "transfer_atk_acc": 0.34659652495163557
    },
    {
        "dataset": "math",
        "type": "task",
        "origin_acc": 16.88,
        "origin_prompt": "Figure out the solution for the following mathematical task with {}.",
        "atk_acc": 15.0,
        "atk_prompt": "Figure out the solution for the followinɡ mathematical task with {}. ",
        "transfer_ori_acc": 21.25,
        "transfer_atk_acc": 21.875
    },
    {
        "dataset": "math",
        "type": "role",
        "origin_acc": 16.88,
        "origin_prompt": "Pretend you are a math teacher and solve this mathematical question concerning {}.",
        "atk_acc": 13.75,
        "atk_prompt": "Pretend you are a math teacher and settle this matehmatical question on {}. ",
        "transfer_ori_acc": 20.625,
        "transfer_atk_acc": 18.125
    }
]