Learning Natural Language Inference with LSTM

Abstract

Natural language inference (NLI) is a fundamentally important task in natural language processing that has many applications. The recently released Stanford Natural Language Inference (SNLI) corpus has made it possible to develop and evaluate learning-centered methods such as deep neural networks for natural language inference (NLI). In this paper, we propose a special long short-term memory (LSTM) architecture for NLI. Our model builds on top of a recently proposed neural attention model for NLI but is based on a significantly different idea. Instead of deriving sentence embeddings for the premise and the hypothesis to be used for classification, our solution uses a match-LSTM to perform word-by-word matching of the hypothesis with the premise. This LSTM is able to place more emphasis on important word-level matching results. In particular, we observe that this LSTM remembers important mismatches that are critical for predicting the contradiction or the neutral relationship label. On the SNLI corpus, our model achieves an accuracy of 86.1%, outperforming the state of the art.

10 pages, 2 figures

References in corpus (2)

• Order-Embeddings of Images and Language
• Natural Language Inference by Tree-Based Convolution and Heuristic Matching

Cited by in corpus (26)

• Recent Advances in Recurrent Neural Networks
• Bilateral Multi-Perspective Matching for Natural Language Sentences
• Conditional Variational Autoencoder for Neural Machine Translation
• End-to-End Answer Chunk Extraction and Ranking for Reading Comprehension
• Machine Reading Comprehension: a Literature Review
• Enhance word representation for out-of-vocabulary on Ubuntu dialogue corpus
• A Parallel-Hierarchical Model for Machine Comprehension on Sparse Data
• Paraphrase Generation with Deep Reinforcement Learning
• Using LSTM and SARIMA Models to Forecast Cluster CPU Usage
• Improving Natural Language Inference Using External Knowledge in the Science Questions Domain
• Explicit Interaction Model towards Text Classification
• Ranking Paragraphs for Improving Answer Recall in Open-Domain Question Answering
• Can Neural Networks Understand Logical Entailment?
• Label-aware Document Representation via Hybrid Attention for Extreme Multi-Label Text Classification
• Sequential Matching Network: A New Architecture for Multi-turn Response Selection in Retrieval-based Chatbots
• On the Effective Use of Pretraining for Natural Language Inference
• A Sequential Matching Framework for Multi-turn Response Selection in Retrieval-based Chatbots
• Looking Beyond Sentence-Level Natural Language Inference for Downstream Tasks
• Cross-Attention End-to-End ASR for Two-Party Conversations
• Syntax-based Attention Model for Natural Language Inference
• What If We Simply Swap the Two Text Fragments? A Straightforward yet Effective Way to Test the Robustness of Methods to Confounding Signals in Nature Language Inference Tasks
• Why and How to Pay Different Attention to Phrase Alignments of Different Intensities
• Quantized Neural Network Inference with Precision Batching
• AWE: Asymmetric Word Embedding for Textual Entailment
• Exploring Lexical Irregularities in Hypothesis-Only Models of Natural Language Inference
• Improving Multi-Turn Response Selection Models with Complementary Last-Utterance Selection by Instance Weighting