Towards a Deep and Unified Understanding of Deep Neural Models in NLP

作者：Chaoyu Guan, Xiting Wang, Quanshi Zhang, Runjin Chen, Di He, Xing Xie · 发表于：International Conference on Machine Learning · 年份：2019 · 被引用次数：73 · 研究领域：Topic Modeling、Explainable Artificial Intelligence (XAI)、Natural Language Processing Techniques

We define a unified information-based measure to provide quantitative explanations on how intermediate layers of deep Natural Language Processing (NLP) models leverage information of input words. Our method advances existing explanation methods by addressing issues in coherency and generality. Explanations generated by using our method are consistent and faithful across different timestamps, layers, and models. We show how our method can be applied to four widely used models in NLP and explain their performances on three real-world benchmark datasets.