Understanding Syllogistic Reasoning in LLMs from Formal and Natural Language Perspectives

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Poddar, Aheli, Sahoo, Saptarshi, Ghosh, Sujata
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909977369640960
author Poddar, Aheli
Sahoo, Saptarshi
Ghosh, Sujata
author_facet Poddar, Aheli
Sahoo, Saptarshi
Ghosh, Sujata
contents We study syllogistic reasoning in LLMs from the logical and natural language perspectives. In process, we explore fundamental reasoning capabilities of the LLMs and the direction this research is moving forward. To aid in our studies, we use 14 large language models and investigate their syllogistic reasoning capabilities in terms of symbolic inferences as well as natural language understanding. Even though this reasoning mechanism is not a uniform emergent property across LLMs, the perfect symbolic performances in certain models make us wonder whether LLMs are becoming more and more formal reasoning mechanisms, rather than making explicit the nuances of human reasoning.
format Preprint
id arxiv_https___arxiv_org_abs_2512_12620
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Understanding Syllogistic Reasoning in LLMs from Formal and Natural Language Perspectives
Poddar, Aheli
Sahoo, Saptarshi
Ghosh, Sujata
Computation and Language
Artificial Intelligence
I.2.7; I.2.3; F.4.1
We study syllogistic reasoning in LLMs from the logical and natural language perspectives. In process, we explore fundamental reasoning capabilities of the LLMs and the direction this research is moving forward. To aid in our studies, we use 14 large language models and investigate their syllogistic reasoning capabilities in terms of symbolic inferences as well as natural language understanding. Even though this reasoning mechanism is not a uniform emergent property across LLMs, the perfect symbolic performances in certain models make us wonder whether LLMs are becoming more and more formal reasoning mechanisms, rather than making explicit the nuances of human reasoning.
title Understanding Syllogistic Reasoning in LLMs from Formal and Natural Language Perspectives
topic Computation and Language
Artificial Intelligence
I.2.7; I.2.3; F.4.1
url https://arxiv.org/abs/2512.12620