arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2008.03945cs.CL
收藏

BERT中解决常识问题的常识线索研究

On Commonsense Cues in BERT for Solving Commonsense Tasks

Fudan University(复旦大学);Microsoft Research Asia(微软亚洲研究院);Westlake University(西湖大学);Zhejiang University(浙江大学)

Leyang Cui, Sijie Cheng, Yu Wu, Yue Zhang

摘要:

BERT已被用于解决常识性的任务,如CommonsenseQA。虽然先前的研究已经发现BERT在某种程度上确实包含常识信息,但已经有工作表明,预先训练的模型可以依赖于虚假的关联(例如,数据偏差)而不是解决情感分类和其他问题的关键线索。我们定量地研究了BERT在解决常识性任务时结构性常识线索的存在,以及这些线索对模型预测的重要性。使用两种不同的测量方法,我们发现BERT确实使用相关知识来解决任务,并且常识知识的存在与模型精度呈正相关。

英文摘要:

BERT has been used for solving commonsense tasks such as CommonsenseQA. While prior research has found that BERT does contain commonsense information to some extent, there has been work showing that pre-trained models can rely on spurious associations (e.g., data bias) rather than key cues in solving sentiment classification and other problems. We quantitatively investigate the presence of structural commonsense cues in BERT when solving commonsense tasks, and the importance of such cues for the model prediction. Using two different measures, we find that BERT does use relevant knowledge for solving the task, and the presence of commonsense knowledge is positively correlated to the model accuracy.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑