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 2110.06223cs.CLcs.AI
收藏

自然语言解释对Few-Shot NLI中分布外泛化的影响

Investigating the Effect of Natural Language Explanations on Out-of-Distribution Generalization in Few-shot NLI

University of Chicago(芝加哥大学)

Yangqiaoyu Zhou, Chenhao Tan

摘要:

虽然神经模型在SNLI等数据集中表现出了很强的性能,但它们缺乏泛化分布外(OOD)的能力。在这项工作中,我们制定了一个少样本学习设置和检查的影响,自然语言解释对面向对象的推广。我们利用HANS数据集中的模板,并为每个模板构建模板化的自然语言解释。尽管生成的解释显示出与地面实况解释相比具有竞争力的BLEU分数,但它们未能提高预测性能。我们进一步表明,生成的解释往往产生幻觉的信息和错过的关键元素,表明标签。

英文摘要:

Although neural models have shown strong performance in datasets such as SNLI, they lack the ability to generalize out-of-distribution (OOD). In this work, we formulate a few-shot learning setup and examine the effects of natural language explanations on OOD generalization. We leverage the templates in the HANS dataset and construct templated natural language explanations for each template. Although generated explanations show competitive BLEU scores against groundtruth explanations, they fail to improve prediction performance. We further show that generated explanations often hallucinate information and miss key elements that indicate the label.

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。

↑