发表机构
Sun Yat-sen University; Tsinghua Shenzhen International Graduate School, Tsinghua University（中山大学; 清华大学深圳国际研究生院）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
EvoSCM 为科学智能体配备可演化的结构因果模型，通过封闭发现循环修正信念，在 DiscoverPhysics 基准上优于基线，提升科学发现的准确性与实验交互效率。
AI 中文摘要
智能体不仅要学习如何推理，还要学习该相信什么。然而，现有的大语言模型（LLM）智能体通常以自由文本形式表达科学假设，导致其信念隐含，难以测试或修正。我们提出 EvoSCM，该模型为科学智能体配备显式结构因果模型（SCM），其会随着新实验证据的收集而演化。EvoSCM 维护一组相互竞争的 SCM 假设，每个假设编码对环境的候选因果解释，并通过封闭发现循环对其进行演化。每一轮中，智能体从积累的证据中溯因推理出潜在机制，设计具有区分性的干预措施，并承诺可证伪的预测，通过实验对这些预测进行测试。预测与观察之间的差异被归纳提炼为修正规则，以修正每个假设的因果结构和机制，随后智能体会演绎验证修正后的假设集，以匹配积累的证据和结构一致性，从而指导下一轮操作。我们在 DiscoverPhysics 上对 EvoSCM 进行评估，该基准要求智能体通过实验揭示非经典物理世界的隐藏动态。EvoSCM 在科学发现上始终优于基线，产生更准确的解释和预测，同时更有效地利用实验交互。
英文摘要
Scientific discovery depends on the ability to form hypotheses, test them through experiments, and revise them when evidence disagrees. Existing LLM agents support this process by improving their reasoning or actions, but their scientific beliefs are often scattered across free-form reasoning and difficult to update coherently. This makes it difficult to identify what failed, what should change, and whether revisions remain consistent with prior evidence. We introduce EvoSCM, which represents scientific beliefs as a population of structural causal model (SCM) hypotheses that can be tested and revised across experiments. EvoSCM formulates scientific discovery as a closed loop in which causal hypotheses guide experimentation and experimental outcomes drive causal model evolution. Competing SCM hypotheses make falsifiable predictions and guide discriminative experiments that separate alternative explanations. When observations contradict these predictions, EvoSCM distills discrepancies into correction rules identifying which aspects of the hypotheses fail to explain the evidence. These rules guide revisions to causal dependencies, latent factors, mechanisms, and parameters. Revised hypotheses are validated against accumulated evidence and carried forward to guide subsequent experiments, allowing scientific beliefs to evolve cumulatively. We evaluate EvoSCM across physics, chemistry and materials, and biology. It consistently outperforms baseline agents and existing evolution methods, yielding more accurate explanations and predictions with more effective use of experimental budgets. The evolved SCMs also transfer across base models, suggesting reusable scientific knowledge beyond any single model's reasoning process.
