Shortcut-Stacked Sentence Encoders for Multi-Domain Inference

Yixin Nie, Mohit Bansal
8/7/2017
cs.CLcs.AIcs.LG

Official code: easonnie/multiNLI_encoder (58 stars, last commit 2018-05-25).

Abstract

We present a simple sequential sentence encoder for multi-domain natural language inference. Our encoder is based on stacked bidirectional LSTM-RNNs with shortcut connections and fine-tuning of word embeddings. The overall supervised model uses the above encoder to encode two input sentences into two vectors, and then uses a classifier over the vector combination to label the relationship between these two sentences as that of entailment, contradiction, or neural. Our Shortcut-Stacked sentence encoders achieve strong improvements over existing encoders on matched and mismatched multi-domain natural language inference (top non-ensemble single-model result in the EMNLP RepEval 2017 Shared Task (Nangia et al., 2017)). Moreover, they achieve the new state-of-the-art encoding result on the original SNLI dataset (Bowman et al., 2015).

View on arXivView PDF

Code Implementations(1)

easonnie/multiNLI_encoderOfficial100%

This is a repo for multiNLI_encoder.

586Shell, PythonJun 15, 20178 years ago

Similar Papers

Supervised Learning of Universal Sentence Representations from Natural Language Inference Data

80% match2 repos2.3k2017

E-CONAN (Entailment, CONtradition And Neutral) Benchmarks: Arabic Textual Entailment and Natural Inference Datasets

80% match2026

Massively Multilingual Sentence Embeddings for Zero-Shot Cross-Lingual Transfer and Beyond

80% match1 repo3.7k2018

SimCSE: Simple Contrastive Learning of Sentence Embeddings

79% match1 repo3.7k2021

Multilingual Sentence-T5: Scalable Sentence Encoders for Multilingual Applications

79% match2024

Learning General Purpose Distributed Sentence Representations via Large Scale Multi-task Learning

79% match2 repos2.1k2018

Cite this paper

@article{nie2017shortcutstacked,
  title  = {Shortcut-Stacked Sentence Encoders for Multi-Domain Inference},
  author = {Yixin Nie and Mohit Bansal},
  year   = {2017},
  eprint = {1708.02312},
  archivePrefix = {arXiv},
  url    = {https://arxiv.org/abs/1708.02312}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/1708.02312.svg)](https://codeofpaper.com/paper/1708.02312)

Discussion