English

Natural Language Inference from Multiple Premises

Computation and Language 2017-10-10 v1

Abstract

We define a novel textual entailment task that requires inference over multiple premise sentences. We present a new dataset for this task that minimizes trivial lexical inferences, emphasizes knowledge of everyday events, and presents a more challenging setting for textual entailment. We evaluate several strong neural baselines and analyze how the multiple premise task differs from standard textual entailment.

Keywords

Cite

@article{arxiv.1710.02925,
  title  = {Natural Language Inference from Multiple Premises},
  author = {Alice Lai and Yonatan Bisk and Julia Hockenmaier},
  journal= {arXiv preprint arXiv:1710.02925},
  year   = {2017}
}

Comments

Accepted at IJCNLP 2017

R2 v1 2026-06-22T22:07:10.985Z