Implementing Natural Language Inference for comparatives

Authors

  • Izumi Haruta Ochanomizu University, Japan
  • Koji Mineshima Keio University, Japan
  • Daisuke Bekki Ochanomizu University, Japan

Keywords:

comparatives, compositional semantics, theorem proving, Combinatory Categorial Grammar, Natural Language Inference

Abstract

This paper presents a computational framework for Natural Language Inference (NLI) using logic-based semantic representations and theorem-proving. We focus on logical inferences with comparatives and other related constructions in English, which are known for their structural complexity and difficulty in performing efficient reasoning. Using the so-called A-not-A analysis of comparatives, we implement a fully automated system to map various comparative constructions to semantic representations in typed first-order logic via Combinatory Categorial Grammar parsers and to prove entailment relations via a theorem prover. We evaluate the system on a variety of NLI benchmarks that contain challenging inferences, in comparison with other recent logic-based systems and neural NLI models.

DOI:

https://doi.org/10.15398/jlm.v10i1.294

Full article

Published

2023-01-05

How to Cite

Haruta, I., Mineshima, K., & Bekki, D. (2023). Implementing Natural Language Inference for comparatives. Journal of Language Modelling, 10(1), 139–191. https://doi.org/10.15398/jlm.v10i1.294

Issue

Section

Special Section on the Interaction between Formal and Computational Linguistics