Natural Language Inference over Interaction Space

Abstract

Natural Language Inference (NLI) task requires an agent to determine the logical relationship between a natural language premise and a natural language hypothesis. We introduce Interactive Inference Network (IIN), a novel class of neural network architectures that is able to achieve high-level understanding of the sentence pair by hierarchically extracting semantic features from interaction space. We show that an interaction tensor (attention weight) contains semantic information to solve natural language inference, and a denser interaction tensor contains richer semantic information. One instance of such architecture, Densely Interactive Inference Network (DIIN), demonstrates the state-of-the-art performance on large scale NLI copora and large-scale NLI alike corpus. It's noteworthy that DIIN achieve a greater than 20% error reduction on the challenging Multi-Genre NLI (MultiNLI) dataset with respect to the strongest published system.

arXiv comment15 pages, 2 figures, under review as ICLR proceeding, Published at Sixth International Conference on Learning Representations, ICLR 2018

References in corpus (12)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems
• ADADELTA: An Adaptive Learning Rate Method
• Character-level Convolutional Networks for Text Classification
• Enhanced LSTM for Natural Language Inference
• Convolutional Neural Networks for Sentence Classification
• Reasoning about Entailment with Neural Attention
• Learning Natural Language Inference using Bidirectional LSTM model and Inner-Attention
• Bilateral Multi-Perspective Matching for Natural Language Sentences
• Reading Wikipedia to Answer Open-Domain Questions
• Shortcut-Stacked Sentence Encoders for Multi-Domain Inference
• Recurrent Neural Network-Based Sentence Encoder with Gated Attention for Natural Language Inference

Cited by in corpus (52)

• BERTScore: Evaluating Text Generation with BERT
• MolTrans: Molecular Interaction Transformer for Drug Target Interaction Prediction
• GLUE: A Multi-Task Benchmark and Analysis Platform for Natural Language Understanding
• Neural Natural Language Inference Models Enhanced with External Knowledge
• XNLI: Evaluating Cross-lingual Sentence Representations
• Glyce: Glyph-vectors for Chinese Character Representations
• Neural Network Models for Paraphrase Identification, Semantic Textual Similarity, Natural Language Inference, and Question Answering
• Generating Persona Consistent Dialogues by Exploiting Natural Language Inference
• End-to-End Retrieval in Continuous Space
• Adversarial NLI: A New Benchmark for Natural Language Understanding
• Dynamic Integration of Background Knowledge in Neural NLU Systems
• Stochastic Answer Networks for Natural Language Inference
• Semantic Sentence Matching with Densely-connected Recurrent and Co-attentive Information
• Pair-Level Supervised Contrastive Learning for Natural Language Inference
• Combining Fact Extraction and Verification with Neural Semantic Matching Networks
• Towards a Robust Deep Neural Network in Texts: A Survey
• Learning Visually-Grounded Semantics from Contrastive Adversarial Samples
• Explaining Deep Neural Networks
• Systematic Generalization: What Is Required and Can It Be Learned?
• Multi-Level Matching and Aggregation Network for Few-Shot Relation Classification
• Improving Personality Consistency in Conversation by Persona Extending
• Simple and Effective Text Matching with Richer Alignment Features
• Explicit Interaction Model towards Text Classification
• Multi-turn Inference Matching Network for Natural Language Inference
• Interpreting Recurrent and Attention-Based Neural Models: a Case Study on Natural Language Inference
• Breaking NLI Systems with Sentences that Require Simple Lexical Inferences
• Analyzing Compositionality-Sensitivity of NLI Models
• Natural Language Inference over Interaction Space: ICLR 2018 Reproducibility Report
• LadRa-Net: Locally-Aware Dynamic Re-read Attention Net for Sentence Semantic Matching
• Dissecting Contextual Word Embeddings: Architecture and Representation
• GeoSQA: A Benchmark for Scenario-based Question Answering in the Geography Domain at High School Level
• Investigating Biases in Textual Entailment Datasets
• AlignVE: Visual Entailment Recognition Based on Alignment Relations
• Selection Bias Explorations and Debias Methods for Natural Language Sentence Matching Datasets
• Retrieve Synonymous keywords for Frequent Queries in Sponsored Search in a Data Augmentation Way
• You Can Do Better! If You Elaborate the Reason When Making Prediction
• Match${}^2$: A Matching over Matching Model for Similar Question Identification
• Modeling Event Background for If-Then Commonsense Reasoning Using Context-aware Variational Autoencoder
• Wasserstein Distance Regularized Sequence Representation for Text Matching in Asymmetrical Domains
• Robustness to Modification with Shared Words in Paraphrase Identification
• Reliable Evaluations for Natural Language Inference based on a Unified Cross-dataset Benchmark
• WikiContradiction: Detecting Self-Contradiction Articles on Wikipedia
• Combining word embeddings and convolutional neural networks to detect duplicated questions
• AWE: Asymmetric Word Embedding for Textual Entailment
• REGMAPR - Text Matching Made Easy
• Multi-task Sentence Encoding Model for Semantic Retrieval in Question Answering Systems
• Mitigating Annotation Artifacts in Natural Language Inference Datasets to Improve Cross-dataset Generalization Ability
• Inducing Alignment Structure with Gated Graph Attention Networks for Sentence Matching
• Pre-trained Language Model Based Active Learning for Sentence Matching
• Generating Contradictory, Neutral, and Entailing Sentences
• GiBERT: Introducing Linguistic Knowledge into BERT through a Lightweight Gated Injection Method
• Discriminatively-Tuned Generative Classifiers for Robust Natural Language Inference