Explaining Text Matching on Neural Natural Language Inference
Explaining Text Matching on Neural Natural Language Inference
复制标题
DOI：
10.1145/3418052
复制
发表时间：
2020-09
期刊：
影响因子：
--
通讯作者：
Youngwoo Kim;Myungha Jang;J. Allan
中科院分区：
文献类型：
--
作者：
Youngwoo Kim;Myungha Jang;J. Allan
自然语言推理（NLI）是检测给定句子对中蕴涵或矛盾的存在的任务。虽然NLI技术可以帮助许多信息检索任务，NLI的大多数解决方案是神经的方法，其缺乏可解释性禁止直接集成和诊断进一步改进。我们的目标是从神经模型中为NLI生成令牌级解释的任务。许多现有的标记级解释方法要么计算成本高，要么需要额外的注释进行训练。在这篇文章中，我们首先介绍了一种新的方法来训练解释生成器，它不需要额外的人类标签。相反，解释生成器的训练目标是预测当部分输入被修改时模型的分类输出将如何变化。其次，我们建议在多任务学习环境中构建一个解释生成器，沿着原始NLI任务，以便解释生成器可以利用模型的内部行为。实验结果表明，所提出的解释生成器优于许多强基线。此外，我们的方法在预测时不需要过多的额外计算，这使得它比性能最好的基线快一个数量级。
Natural language inference (NLI) is the task of detecting the existence of entailment or contradiction in a given sentence pair. Although NLI techniques could help numerous information retrieval tasks, most solutions for NLI are neural approaches whose lack of interpretability prohibits both straightforward integration and diagnosis for further improvement. We target the task of generating token-level explanations for NLI from a neural model. Many existing approaches for token-level explanation are either computationally costly or require additional annotations for training. In this article, we first introduce a novel method for training an explanation generator that does not require additional human labels. Instead, the explanation generator is trained with the objective of predicting how the model’s classification output will change when parts of the inputs are modified. Second, we propose to build an explanation generator in a multi-task learning setting along with the original NLI task so the explanation generator can utilize the model’s internal behavior. The experiment results suggest that the proposed explanation generator outperforms numerous strong baselines. In addition, our method does not require excessive additional computation at prediction time, which renders it an order of magnitude faster than the best-performing baseline.