PapersSelected73°

EvoSCM Scientific Belief Revision System

EvoSCM: Scientific Belief Revision Through Causal Model Evolution and Experimentation

Why we picked it

EvoSCM improves scientific reasoning accuracy for researchers working with complex physical systems through structured hypothesis evolution.

EvoSCM introduces explicit structural causal models for scientific agents to evolve hypotheses through experimentation.

Body

The system maintains competing SCM hypotheses that encode candidate causal explanations of environments.Evaluated on DiscoverPhysics benchmark, EvoSCM improves scientific discovery accuracy and prediction effectiveness over baselines.

The agent uses abduction, intervention design, and falsifiable predictions to refine causal structures.

Original · arXiv cs.AI

EvoSCM: Scientific Belief Revision Through Causal Model Evolution and Experimentation

Scientific agents must learn not only how to reason, but also what to believe. However, existing LLM agents typically express scientific hypotheses in free-form text, leaving their beliefs implicit and difficult to test or revise. We introduce EvoSCM, which equips scientific agents with explicit structural causal models that evolve as new experimental evidence is collected. EvoSCM maintains a population of competing SCM hypotheses, each encoding a candidate causal explanation of the environment, and evolves them through a closed discovery loop. In each round, the agent abduces latent mechanisms from accumulated evidence, designs discriminative interventions, and commits to falsifiable predictions that it tests through experimentation. Discrepancies between prediction and observation are inductively distilled into correction rules that revise the causal structures and mechanisms of each hypothesis, and the agent then deductively validates the revised population against accumulated evidence and structural consistency to guide the next round. We evaluate EvoSCM on DiscoverPhysics, a benchmark requiring agents to uncover the hidden dynamics of noncanonical physical worlds through experimentation. EvoSCM consistently improves scientific discovery over baselines, yielding more accurate explanations and predictions while making more effective use of experimental interactions.

Industry Coordinates & Source Verification

This report is categorized under AITOP's "Papers" desk. Key entities and technical topics under continuous tracking include EvoSCM, causal models, scientific agents, DiscoverPhysics.

Source attribution: curated from arXiv cs.AI (@Qing Zhao, Haowei Li, Weijian Deng, Pengxu Wei, Liang Lin).