Explaining Textual Entailment with Lexical Entailments: Using LLMs to Supply Lexical Relations for Formal Proofs

2026-09-26

AI Summary

This study investigates whether large language models (LLMs) can supply sufficient lexical entailment knowledge to support formal proofs in logical natural language inference (NLI) systems. It introduces a novel task of explaining sentence-level entailment through lexical entailment relations and constructs a dedicated dataset to evaluate the capacity of LLMs to generate structured lexical explanations. These explanations are further validated within a neuro-symbolic framework integrated with the LangPro theorem prover to assess their contribution to formal reasoning. Experimental results demonstrate that this task remains highly challenging: the entailment relations generated by LLMs are only partially reliable, frequently restricted to specific instances, and offer limited substantive support for theorem proving. These findings reveal fundamental limitations of current LLMs in performing rigorous logical reasoning.

Abstract

Large Language Models (LLMs) are highly capable of natural language reasoning and appear to store a great deal of lexical knowledge, but it is still unclear how much of this knowledge they actually use when reasoning, and whether they use it in the right way. On the other hand, logic-based Natural Language Inference (NLI) systems provide transparent and formally grounded reasoning, but they need to be supplied with rich lexical knowledge to prove inferences beyond purely logical ones. In this paper, we evaluate whether LLMs can identify all lexical knowledge needed to solve NLI problems and how much this knowledge contributes to proof search in a logic-based NLI system. Our research focuses exclusively on structured lexical entailments (e.g., chinchilla$\sqsubseteq$small animal) as a proxy for structured explanations for NLI problems with an entailment label. First, we curate a dataset for a new task of explaining sentential entailments with a set of lexical entailments. The dataset is used to intrinsically evaluate LLMs on generating structured lexical explanations. Then, we use NLI as an extrinsic evaluation in a simple neuro-symbolic setting, assessing whether LLMs can supply sufficient lexical relations to LangPro, a natural-logic theorem prover for natural language. The results show that the proposed task remains challenging even for hosted proprietary LLMs, and that their contribution to theorem proving is moderate: generated relations are often only partially sound and may be tailored to the specific NLI problem rather than representing generally valid lexical knowledge.

Problem

• Natural Language Inference
• Lexical Entailment
• Large Language Models
• Theorem Proving
• Neuro-symbolic

Innovation

• Large Language Models
• Natural Language Inference
• Neuro-symbolic
• Lexical Entailment
• Theorem Proving