Implementing Natural Language Inference for comparatives

Izumi Haruta, Koji Mineshima, Daisuke Bekki

Research output: Contribution to journalArticlepeer-review


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.

Original languageEnglish
Pages (from-to)139-191
Number of pages53
JournalJournal of Language Modelling
Issue number1
Publication statusPublished - 2022


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

ASJC Scopus subject areas

  • Modelling and Simulation
  • Linguistics and Language
  • Computer Science Applications


Dive into the research topics of 'Implementing Natural Language Inference for comparatives'. Together they form a unique fingerprint.

Cite this