Abductive Reasoning with Probabilistic Commonsense

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Cotnareanu, Joseph, Roverato, Chiara, Zhou, Han, Chetelat, Didier, Zhang, Yingxue, Coates, Mark
Formato: Preprint
Publicado: 2026
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866914545346281472
author Cotnareanu, Joseph
Roverato, Chiara
Zhou, Han
Chetelat, Didier
Zhang, Yingxue
Coates, Mark
author_facet Cotnareanu, Joseph
Roverato, Chiara
Zhou, Han
Chetelat, Didier
Zhang, Yingxue
Coates, Mark
contents Recent efforts to improve the reasoning abilities of Large Language Models (LLMs) have focused on integrating formal logic solvers within neurosymbolic frameworks. A key challenge is that formal solvers lack commonsense world knowledge, preventing them from making reasoning steps that humans find obvious. Prior methods address this by using LLMs to supply missing commonsense assumptions, but these approaches implicitly assume universal agreement on such commonsense facts. In reality, commonsense beliefs vary across individuals. We propose a probabilistic framework for abductive commonsense reasoning that explicitly models this variation, aiming to determine whether most people would judge a statement as true or false. We introduce Probabilistic Abductive CommonSense (PACS), a novel algorithm that uses an LLM and a formal solver to sample proofs as observations of individuals' distinct commonsense beliefs, and aggregates conclusions across these samples. Empirically, PACS outperforms chain-of-thought reasoning, prior neurosymbolic methods, and search-based approaches across multiple benchmarks.
format Preprint
id arxiv_https___arxiv_org_abs_2605_08011
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Abductive Reasoning with Probabilistic Commonsense
Cotnareanu, Joseph
Roverato, Chiara
Zhou, Han
Chetelat, Didier
Zhang, Yingxue
Coates, Mark
Artificial Intelligence
Computation
Recent efforts to improve the reasoning abilities of Large Language Models (LLMs) have focused on integrating formal logic solvers within neurosymbolic frameworks. A key challenge is that formal solvers lack commonsense world knowledge, preventing them from making reasoning steps that humans find obvious. Prior methods address this by using LLMs to supply missing commonsense assumptions, but these approaches implicitly assume universal agreement on such commonsense facts. In reality, commonsense beliefs vary across individuals. We propose a probabilistic framework for abductive commonsense reasoning that explicitly models this variation, aiming to determine whether most people would judge a statement as true or false. We introduce Probabilistic Abductive CommonSense (PACS), a novel algorithm that uses an LLM and a formal solver to sample proofs as observations of individuals' distinct commonsense beliefs, and aggregates conclusions across these samples. Empirically, PACS outperforms chain-of-thought reasoning, prior neurosymbolic methods, and search-based approaches across multiple benchmarks.
title Abductive Reasoning with Probabilistic Commonsense
topic Artificial Intelligence
Computation
url https://arxiv.org/abs/2605.08011