arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2104.08142cs.CLcs.LG
收藏

基于人的注意力监督的鲁棒自然语言推理

Supervising Model Attention with Human Explanations for Robust Natural Language Inference

Imperial College London(帝国理工学院);Technion – Israel Institute of Technology(以色列理工学院)

Joe Stacey, Yonatan Belinkov, Marek Rei

摘要:

众所周知,自然语言推理(NLI)模型可以从其训练数据中的偏差和伪影中学习,从而影响它们对其他不可见数据集的泛化能力。现有的去偏置方法集中在防止模型学习这些偏置,这可能导致限制性模型和较低的性能。相反,我们研究如何教模型人类将接近NLI任务,以学习功能,将更好地推广到以前看不见的例子。使用自然语言解释,我们监督模型的注意力权重,以鼓励更多的注意力集中在解释中出现的单词,显着提高模型的性能。我们的实验表明,这种方法的分布内改进也伴随着分布外改进,监督模型从更好地推广到其他NLI数据集的特征中学习。对模型的分析表明,人类的解释鼓励增加对重要词的关注,更多地关注前提词,而较少关注标点符号和停止词。

英文摘要:

Natural Language Inference (NLI) models are known to learn from biases and artefacts within their training data, impacting how well they generalise to other unseen datasets. Existing de-biasing approaches focus on preventing the models from learning these biases, which can result in restrictive models and lower performance. We instead investigate teaching the model how a human would approach the NLI task, in order to learn features that will generalise better to previously unseen examples. Using natural language explanations, we supervise the model's attention weights to encourage more attention to be paid to the words present in the explanations, significantly improving model performance. Our experiments show that the in-distribution improvements of this method are also accompanied by out-of-distribution improvements, with the supervised models learning from features that generalise better to other NLI datasets. Analysis of the model indicates that human explanations encourage increased attention on the important words, with more attention paid to words in the premise and less attention paid to punctuation and stop-words.

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。

↑