Abductive Reasoning with Probabilistic Commonsense
Joseph Cotnareanu, Chiara Roverato, Han Zhou, Didier Chetelat, Yingxue Zhang + 1 more
TLDR
PACS offers a probabilistic framework for abductive commonsense reasoning, explicitly modeling belief variation to enhance LLM reasoning capabilities.
Key contributions
- Proposes Probabilistic Abductive CommonSense (PACS) algorithm for neurosymbolic reasoning.
- Explicitly models variation in commonsense beliefs, unlike prior methods assuming universal agreement.
- Uses LLMs and formal solvers to sample and aggregate individual commonsense proofs.
- Achieves state-of-the-art performance across multiple reasoning benchmarks.
Why it matters
This paper addresses a critical limitation in neurosymbolic AI by modeling the variability of commonsense beliefs, enabling more robust and human-like abductive reasoning. It moves beyond the assumption of universal agreement, significantly advancing AI's capability in complex, ambiguous reasoning tasks.
Original Abstract
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.
📬 Weekly AI Paper Digest
Get the top 10 AI/ML arXiv papers from the week — summarized, scored, and delivered to your inbox every Monday.