Natural Language Inference by Tree-Based Convolution and Heuristic Matching
arXiv:1512.08422
Abstract
In this paper, we propose the TBCNN-pair model to recognize entailment and contradiction between two sentences. In our model, a tree-based convolutional neural network (TBCNN) captures sentence-level semantics; then heuristic matching layers like concatenation, element-wise product/difference combine the information in individual sentences. Experimental results show that our model outperforms existing sentence encoding-based approaches by a large margin.
Accepted by ACL'16 as a short paper
References in corpus (6)
- Convolutional Neural Network Architectures for Matching Natural Language Sentences
- A Convolutional Neural Network for Modelling Sentences
- Improved Relation Classification by Deep Recurrent Neural Networks with Data Augmentation
- Order-Embeddings of Images and Language
- Discriminative Neural Sentence Modeling by Tree-Based Convolution
- StalemateBreaker: A Proactive Content-Introducing Approach to Automatic Human-Computer Conversation
Cited by in corpus (13)
- Learning Natural Language Inference using Bidirectional LSTM model and Inner-Attention
- How Transferable are Neural Networks in NLP Applications?
- Distance-based Self-Attention Network for Natural Language Inference
- Learning Natural Language Inference with LSTM
- A Decomposable Attention Model for Natural Language Inference
- Stochastic Answer Networks for Natural Language Inference
- Shortcut-Stacked Sentence Encoders for Multi-Domain Inference
- Time-Aware Evidence Ranking for Fact-Checking
- Syntax-based Attention Model for Natural Language Inference
- Hierarchical RNN with Static Sentence-Level Attention for Text-Based Speaker Change Detection
- SNU_IDS at SemEval-2018 Task 12: Sentence Encoder with Contextualized Vectors for Argument Reasoning Comprehension
- Question-Aware Sentence Gating Networks for Question and Answering
- Sequential Sentence Matching Network for Multi-turn Response Selection in Retrieval-based Chatbots