When Should an AI Scientist Stop? Verifiable Experiment Steering and Refusal for Autonomous Discovery

2026-05-26 · Neel Tushar Shah, Manglam Kartik arxiv

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

📄 PDF Abstract BibTeX arXiv:2606.07576

Code (0)

등록된 구현이 없습니다.

Providing Assurance and Scrutability on Shared Data and Machine Learning Models with Verifiable Credentials

2021-05-13 · Iain Barclay, Alun Preece, Ian Taylor, Swapna K. Radha 외

Adopting shared data resources requires scientists to place trust in the originators of the data. When shared data is later used in the development of artificial intelligence (AI) systems or machine learning (ML) models,…

BIG-bench Machine Learning

Should I Stop or Should I Go: Early Stopping with Heterogeneous Populations

2023-09-21 · NeurIPS 2023 11

Randomized experiments often need to be stopped prematurely due to the treatment having an unintended harmful effect. Existing methods that determine when to stop an experiment early are typically applied to the data in …

AutoResearchClaw: Self-Reinforcing Autonomous Research with Human-AI Collaboration

2026-05-19 · Jiaqi Liu, Shi Qiu, Mairui Li, Bingzhou Li 외 arxiv

Automating scientific discovery requires more than generating papers from ideas. Real research is iterative: hypotheses are challenged from multiple perspectives, experiments fail and inform the next attempt, and lessons…

LLM Social Simulations Are a Promising Research Method

2025-04-03 · Jacy Reese Anthis, Ryan Liu, Sean M. Richardson, Austin C. Kozlowski 외

Accurate and verifiable large language model (LLM) simulations of human research subjects promise an accessible data source for understanding human behavior and training new AI systems. However, results to date have been…

Language Modeling Language Modelling Large Language Model Marketing+1

Next integrated result modelling for stopping the text field recognition process in a video using a result model with per-character alternatives

2019-10-09 · Konstantin Bulatov, Boris Savelyev, Vladimir V. Arlazarov

In the field of document analysis and recognition using mobile devices for capturing, and the field of object recognition in a video stream, an important problem is determining the time when the capturing process should …

Clustering Object Recognition