Stochastic Answer Networks for Natural Language Inference
arXiv:1804.07888
Abstract
We propose a stochastic answer network (SAN) to explore multi-step inference strategies in Natural Language Inference. Rather than directly predicting the results given the inputs, the model maintains a state and iteratively refines its predictions. Our experiments show that SAN achieves the state-of-the-art results on three benchmarks: Stanford Natural Language Inference (SNLI) dataset, MultiGenre Natural Language Inference (MultiNLI) dataset and Quora Question Pairs dataset.
6 pages, 1 figures
References in corpus (3)
Cited by in corpus (15)
- Multi-Task Deep Neural Networks for Natural Language Understanding
- Stochastic Answer Networks for Machine Reading Comprehension
- GLoMo: Unsupervisedly Learned Relational Graphs as Transferable Representations
- Stochastic Answer Networks for SQuAD 2.0
- The Microsoft Toolkit of Multi-Task Deep Neural Networks for Natural Language Understanding
- Simple and Effective Text Matching with Richer Alignment Features
- Multi-task Learning with Sample Re-weighting for Machine Reading Comprehension
- Improving Textual Network Embedding with Global Attention via Optimal Transport
- 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
- Matching Text with Deep Mutual Information Estimation
- Inducing Alignment Structure with Gated Graph Attention Networks for Sentence Matching
- Wasserstein Distance Regularized Sequence Representation for Text Matching in Asymmetrical Domains
- Answer Generation through Unified Memories over Multiple Passages
- Conflict as an Inverse of Attention in Sequence Relationship
- Multi-Perspective Inferrer: Reasoning Sentences Relationship from Holistic Perspective