Interpreting Recurrent and Attention-Based Neural Models: a Case Study on Natural Language Inference

Deep learning models have achieved remarkable success in natural language inference (NLI) tasks. While these models are widely explored, they are hard to interpret and it is often unclear how and why they actually work. In this paper, we take a step toward explaining such deep learning based models through a case study on a popular neural model for NLI. In particular, we propose to interpret the intermediate layers of NLI models by visualizing the saliency of attention and LSTM gating signals. We present several examples for which our methods are able to reveal interesting insights and identify the critical information contributing to the model decisions.

[1]  Christopher Potts,et al.  A large annotated corpus for learning natural language inference , 2015, EMNLP.

[2]  Xiaoli Z. Fern,et al.  DR-BiLSTM: Dependent Reading Bidirectional LSTM for Natural Language Inference , 2018, NAACL.

[3]  Jian Zhang,et al.  Natural Language Inference over Interaction Space , 2017, ICLR.

[4]  Zhen-Hua Ling,et al.  Enhanced LSTM for Natural Language Inference , 2016, ACL.

[5]  Jakob Uszkoreit,et al.  A Decomposable Attention Model for Natural Language Inference , 2016, EMNLP.

[6]  Klaus-Robert Müller,et al.  Explaining Recurrent Neural Network Predictions in Sentiment Analysis , 2017, WASSA@EMNLP.

[7]  Andrew Zisserman,et al.  Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Maps , 2013, ICLR.

[8]  Xiaoli Z. Fern,et al.  Dependent Gated Reading for Cloze-Style Question Answering , 2018, COLING.

[9]  Anders Søgaard,et al.  Zero-Shot Sequence Labeling: Transferring Knowledge from Sentences to Tokens , 2018, NAACL.

[10]  Misha Denil,et al.  Modelling, Visualising and Summarising Documents with a Single Convolutional Neural Network , 2014, ArXiv.

[11]  Yoshua Bengio,et al.  Neural Machine Translation by Jointly Learning to Align and Translate , 2014, ICLR.

[12]  Xinlei Chen,et al.  Visualizing and Understanding Neural Models in NLP , 2015, NAACL.

[13]  Luke S. Zettlemoyer,et al.  Deep Contextualized Word Representations , 2018, NAACL.

[14]  Zhiguo Wang,et al.  Bilateral Multi-Perspective Matching for Natural Language Sentences , 2017, IJCAI.

[15]  Daniel Jurafsky,et al.  Understanding Neural Networks through Representation Erasure , 2016, ArXiv.

[16]  Siu Cheung Hui,et al.  A Compare-Propagate Architecture with Alignment Factorization for Natural Language Inference , 2017, ArXiv.

[17]  Ruslan Salakhutdinov,et al.  Gated-Attention Readers for Text Comprehension , 2016, ACL.

[18]  Richard Socher,et al.  Learned in Translation: Contextualized Word Vectors , 2017, NIPS.