LOGicalThought: Logic-Based Ontological Grounding of LLMs for High-Assurance Reasoning

2025-10-01

AI Summary

In high-assurance domains such as law and medicine, defeasible rules (e.g., “P holds generally, but exception Q invalidates P”) pose significant reasoning challenges: a single exception can overturn a general conclusion, undermining LLMs’ accuracy and verifiability. To address this, we propose LOGicalThought, a neurosymbolic architecture that constructs a two-tier symbolic graph and logical context to map long-text reasoning into compact, verifiable logical evaluation—explicitly modeling negation, entailment, and defeasible inference. By integrating a high-level logical language, a symbolic reasoning engine, and a large language model, LOGicalThought enables structured inference over rule-exception structures. Evaluated on four cross-domain benchmarks, it achieves an overall accuracy gain of 11.84%, with improvements of 10.2%, 13.2%, and 5.5% on negation, entailment, and defeasible reasoning subtasks, respectively—substantially enhancing rigorous, trustworthy reasoning in high-assurance settings.

Technology Category

• Knowledge Representation and Reasoning: Computational Complexity of Reasoning
• Cognitive Modeling & Cognitive Systems: Conceptual Inference and Reasoning
• Reasoning under Uncertainty: Other Foundations of Reasoning under Uncertainty

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

High-assurance reasoning, particularly in critical domains such as law and medicine, requires conclusions that are accurate, verifiable, and explicitly grounded in evidence. This reasoning relies on premises codified from rules, statutes, and contracts, inherently involving defeasible or non-monotonic logic due to numerous exceptions, where the introduction of a single fact can invalidate general rules, posing significant challenges. While large language models (LLMs) excel at processing natural language, their capabilities in standard inference tasks do not translate to the rigorous reasoning required over high-assurance text guidelines. Core reasoning challenges within such texts often manifest specific logical structures involving negation, implication, and, most critically, defeasible rules and exceptions. In this paper, we propose a novel neurosymbolically-grounded architecture called LOGicalThought (LogT) that uses an advanced logical language and reasoner in conjunction with an LLM to construct a dual symbolic graph context and logic-based context. These two context representations transform the problem from inference over long-form guidelines into a compact grounded evaluation. Evaluated on four multi-domain benchmarks against four baselines, LogT improves overall performance by 11.84% across all LLMs. Performance improves significantly across all three modes of reasoning: by up to +10.2% on negation, +13.2% on implication, and +5.5% on defeasible reasoning compared to the strongest baseline.

Problem

• Addresses high-assurance reasoning challenges in critical domains
• Handles defeasible logic with exceptions in rule-based systems
• Improves LLM performance on negation, implication, and defeasible reasoning

Innovation

• LOGicalThought combines logical language with LLMs
• It constructs dual symbolic graph and logic contexts
• Transforms long guidelines into compact grounded evaluation