模型:

DeepPavlov/roberta-large-winogrande

英文

RoBERTa Large模型在Winogrande上的微调

该模型在Winogrande数据集(XL尺寸)上进行了微调,任务格式为序列分类,即原始的具有相应选项的句子对被分开、混排并独立进行分类。

模型描述

使用和限制

如何使用

训练数据

WinoGrande-XL 被重新格式化如下:

  • 每个句子都分割在 " _ " 的占位符上。
  • 每个选项都与分割后的第二部分连接在一起,从而将每个示例转换为两个文本段对。
  • 分别使用 True 和 False 标签标记与正确和错误选项相对应的文本段对。
  • 然后对文本段对进行洗牌。
  • 例如,

    {
      "answer": "2",
      "option1": "plant",
      "option2": "urn",
      "sentence": "The plant took up too much room in the urn, because the _ was small."
    }
    

    变成

    {
      "sentence1": "The plant took up too much room in the urn, because the ",
      "sentence2": "plant was small.",
      "label": false
    }
    

    {
      "sentence1": "The plant took up too much room in the urn, because the ",
      "sentence2": "urn was small.",
      "label": true
    }
    

    然后将这些句子对作为独立的示例处理。

    BibTeX条目和引用信息

    @article{sakaguchi2019winogrande,
        title={WinoGrande: An Adversarial Winograd Schema Challenge at Scale},
        author={Sakaguchi, Keisuke and Bras, Ronan Le and Bhagavatula, Chandra and Choi, Yejin},
        journal={arXiv preprint arXiv:1907.10641},
        year={2019}
    }
    
    @article{DBLP:journals/corr/abs-1907-11692,
      author    = {Yinhan Liu and
                   Myle Ott and
                   Naman Goyal and
                   Jingfei Du and
                   Mandar Joshi and
                   Danqi Chen and
                   Omer Levy and
                   Mike Lewis and
                   Luke Zettlemoyer and
                   Veselin Stoyanov},
      title     = {RoBERTa: {A} Robustly Optimized {BERT} Pretraining Approach},
      journal   = {CoRR},
      volume    = {abs/1907.11692},
      year      = {2019},
      url       = {http://arxiv.org/abs/1907.11692},
      archivePrefix = {arXiv},
      eprint    = {1907.11692},
      timestamp = {Thu, 01 Aug 2019 08:59:33 +0200},
      biburl    = {https://dblp.org/rec/journals/corr/abs-1907-11692.bib},
      bibsource = {dblp computer science bibliography, https://dblp.org}
    }