Paper: Modeling Human Inference Process for Textual Entailment Recognition

ACL ID P13-2079
Title Modeling Human Inference Process for Textual Entailment Recognition
Venue Annual Meeting of the Association of Computational Linguistics
Session Short Paper
Year 2013
Authors

This paper aims at understanding what hu- man think in textual entailment (TE) recogni- tion process and modeling their thinking pro- cess to deal with this problem. We first ana- lyze a labeled RTE-5 test set and find that the negative entailment phenomena are very ef- fective features for TE recognition. Then, a method is proposed to extract this kind of phenomena from text-hypothesis pairs auto- matically. We evaluate the performance of using the negative entailment phenomena on both the English RTE-5 dataset and Chinese NTCIR-9 RITE dataset, and conclude the same findings.