Semantic Sentence Matching with Densely-connected Recurrent and Co-attentive Information

Abstract

Sentence matching is widely used in various natural language tasks such as natural language inference, paraphrase identification, and question answering. For these tasks, understanding logical and semantic relationship between two sentences is required but it is yet challenging. Although attention mechanism is useful to capture the semantic relationship and to properly align the elements of two sentences, previous methods of attention mechanism simply use a summation operation which does not retain original features enough. Inspired by DenseNet, a densely connected convolutional network, we propose a densely-connected co-attentive recurrent neural network, each layer of which uses concatenated information of attentive features as well as hidden features of all the preceding recurrent layers. It enables preserving the original and the co-attentive feature information from the bottommost word embedding layer to the uppermost recurrent layer. To alleviate the problem of an ever-increasing size of feature vectors due to dense concatenation operations, we also propose to use an autoencoder after dense concatenation. We evaluate our proposed architecture on highly competitive benchmark datasets related to sentence matching. Experimental results show that our architecture, which retains recurrent and attentive features, achieves state-of-the-art performances for most of the tasks.

arXiv commentAccepted at AAAI 2019

References in corpus (14)

• Google's Neural Machine Translation System: Bridging the Gap between Human and Machine Translation
• Enhanced LSTM for Natural Language Inference
• A large annotated corpus for learning natural language inference
• Neural Natural Language Inference Models Enhanced with External Knowledge
• Learned in Translation: Contextualized Word Vectors
• Natural Language Inference over Interaction Space
• Bilateral Multi-Perspective Matching for Natural Language Sentences
• Distance-based Self-Attention Network for Natural Language Inference
• Shortcut-Stacked Sentence Encoders for Multi-Domain Inference
• Reinforced Self-Attention Network: a Hybrid of Hard and Soft Attention for Sequence Modeling
• Learning Loss Functions for Semi-supervised Learning via Discriminative Adversarial Networks
• Recurrent Neural Network-Based Sentence Encoder with Gated Attention for Natural Language Inference
• Neural Paraphrase Identification of Questions with Noisy Pretraining
• Learning Context-Sensitive Convolutional Filters for Text Processing

Cited by in corpus (20)

• The Natural Language Decathlon: Multitask Learning as Question Answering
• Multi-Task Deep Neural Networks for Natural Language Understanding
• A Question-Entailment Approach to Question Answering
• Generating Persona Consistent Dialogues by Exploiting Natural Language Inference
• Utilizing BERT Intermediate Layers for Aspect Based Sentiment Analysis and Natural Language Inference
• Fake News Detection as Natural Language Inference
• On the Importance of Delexicalization for Fact Verification
• Multi-Level Matching and Aggregation Network for Few-Shot Relation Classification
• LIMIT-BERT : Linguistic Informed Multi-Task BERT
• Knowledge-Grounded Dialogue Generation with Pre-trained Language Models
• Attention Boosted Sequential Inference Model
• Selection Bias Explorations and Debias Methods for Natural Language Sentence Matching Datasets
• Non-entailed subsequences as a challenge for natural language inference
• Effective Subword Segmentation for Text Comprehension
• 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
• Probing Contextualized Sentence Representations with Visual Awareness
• Robustness to Modification with Shared Words in Paraphrase Identification
• AWE: Asymmetric Word Embedding for Textual Entailment
• Towards Language Agnostic Universal Representations
• BERT-DRE: BERT with Deep Recursive Encoder for Natural Language Sentence Matching