AI科学家实验决策的可信判断框架，能自动识别该继续还是停止探索。
When Should an AI Scientist Stop? Verifiable Experiment Steering and Refusal for Autonomous Discovery
- 基于未解空间的实验选择、模糊性消解与残差检测三机制协同
- 在5个测试中以129胜0平15负击败传统方法，在d=8时显著优越（p<1e-21）
- 可发现并撤销药物代谢机制误判，适合自主科研系统可靠性验证
我们提出CARTOGRAPH，一种面向AI科学家的验证层，结合未解空间实验引导（select）、显式模糊性闭合（resolve）和基于残差的库外不充分性检测（refuse）。在局部线性高斯桥模型下，原始未解投影为各向同性未解费雪信息迹，而CARTOGRAPH-A是精确的未解A最优规则；局部比较器如闭式EIG和Box-Hill并非全局等价物。在五个测试平台中，CARTOGRAPH-A在d=8时以129胜0平15负击败原始投影（p < 10^-21），且在结构化级联中可重复。更关键的是，该框架初步识别出三个库外药代动力学机制，随后因残差暴露结构不适配而撤回识别，而一个扰动内库对照则持续被识别。在低维药代动力学和过滤版EPA设置中，理论预测的近似分歧与实际观测一致。最后，在对40个已发表A-Lab自主材料系统正向结论的回溯审计中，拒识防护机制标记了所有后来被人工复核为不确定的4个声明，同时通过了32/36个确认有效的声明。代码已开源：https://github.com/ai4science-boed/cartograph.git
原文摘要 · Abstract (English)
We present CARTOGRAPH, a verification layer for AI scientists that couples unresolved-subspace experiment steering (select), explicit ambiguity closure (resolve), and residual-based library inadequacy detection (refuse). Under a local linear-Gaussian bridge, raw unresolved projection is the isotropic unresolved Fisher-information trace, while CARTOGRAPH-A is the exact unresolved A-optimal rule; closed-form EIG and Box-Hill arise as local comparators rather than global equivalents. Across five testbeds, CARTOGRAPH-A beats raw projection 129W/0T/15L at d = 8 (p < 10^-21) in a replicated structured cascade. More distinctively, the framework tentatively identifies three out-of-library pharmacokinetic mechanisms and then revokes those identifications as residuals expose structural misfit, while one perturbed in-library control stays identified throughout. In low-dimensional pharmacokinetic and filtered EPA settings, near-ties against disagreement are predicted by theory and observed. Finally, in a retrospective audit of 40 positive claims from the published A-Lab autonomous materials system, the refuse guard flags all 4 claims later marked inconclusive under manual reanalysis while passing 32/36 confirmed claims. Code is available at https://github.com/ai4science-boed/cartograph.git
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。