Epistemic Closure: Autonomous Mechanism Completion for Physically Consistent Simulation

2026-03-10

AI Summary

This work addresses the problem of “physical hallucinations” in large language models when applied to scientific modeling, which often arise from overlooking implicit thermodynamic assumptions—such as undrained conditions. The authors propose a neuro-symbolic generative agent that acts as a cognitive overseer atop numerical solvers, integrating modular constitutive skills, dimensionless scaling analysis (e.g., Deborah number), and chain-of-thought reasoning to automatically identify, validate, and complete hidden physical mechanisms. This approach represents the first demonstration of AI autonomously correcting implicit assumptions in scientific literature, elevating models from mere coding assistants to epistemologically capable scientific partners. In a low-permeability sandstone thermo-poroelasticity case, the agent correctly inferred a drained regime (De << 1), reconstructed the missing dissipation mechanism, and predicted a stable stress path consistent with experimental observations, thereby avoiding an erroneous catastrophic failure prediction.

Technology Category

• Cognitive Modeling & Cognitive Systems: Agent Architectures
• Multiagent Systems: Agent-Based Simulation and Emergent Behavior
• Knowledge Representation and Reasoning: Automated Reasoning and Theorem Proving

Application Category

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

Abstract

The integration of Large Language Models (LLMs) into scientific discovery is currently hindered by the Implicit Context problem, where governing equations extracted from literature contain invisible thermodynamic assumptions (e.g., undrained conditions) that standard generative models fail to recognize. This leads to Physical Hallucination: the generation of syntactically correct solvers that faithfully execute physically invalid laws. Here, we introduce a Neuro-Symbolic Generative Agent that functions as a cognitive supervisor atop traditional numerical engines. By encapsulating physical laws into modular Constitutive Skills and leveraging latent intrinsic priors, the Agent employs a Chain-of-Thought reasoning workflow to autonomously validate, prune, and complete physical mechanisms. We demonstrate this capability on the challenge of thermal pressurization in low-permeability sandstone. While a standard literature-retrieval baseline erroneously predicts catastrophic material failure by blindly adopting a rigid "undrained" simplification, our Agent autonomously identifies the system as operating in a drained regime (Deborah number De << 1) via dimensionless scaling analysis. Consequently, it inductively completes the missing dissipation mechanism (Darcy flow) required to satisfy boundary constraints, predicting a stable stress path consistent with experimental reality. This work establishes a paradigm where AI agents transcend the role of coding assistants to act as epistemic partners, capable of reasoning about and correcting the theoretical assumptions embedded in scientific data.

Problem

• Implicit Context
• Physical Hallucination
• Epistemic Closure
• Constitutive Skills
• Thermodynamic Assumptions

Innovation

• Neuro-Symbolic Generative Agent
• Physical Hallucination
• Constitutive Skills
• Chain-of-Thought Reasoning
• Epistemic Closure