Science sandboxes measure the scientific capability of AI agents

• URL: http://arxiv.org/abs/2608.30165v1
• Date: Mon, 31 Aug 2026 02:33:38 GMT
• Title: Science sandboxes measure the scientific capability of AI agents
• Abstract summary: We introduce science sandboxes, a framework for studying this capability in AI agents.<n>We instantiate this framework in two biological settings, models of regulatory genomics and protein fitness prediction.<n>In particular, their scientific reasoning deteriorated when they encountered systems whose rules fell outside familiar biological priors.
• Score: 1.1846810006828092
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/
• Abstract: Scientific progress depends not only on finding solutions, but on learning the rules that explain why they work and using that understanding to design better experiments. We introduce science sandboxes, a framework for studying this capability in AI agents through repeated cycles of experimentation, feedback, and hypothesis revision. Science sandboxes invite an agent to query the natural world in different ways, ranging from "wet" physical experiments, to "damp" predictive models trained on empirical data, to "dry" invented rules. By establishing a common experimental loop and a protocol for evaluating agents within it, science sandboxes allow assessment of both quantitative performance on specific metrics and qualitative scientific reasoning, across a spectrum of empirical verifiability. Here, we instantiate this framework in two biological settings, models of regulatory genomics and protein fitness prediction, and examine the capabilities of frontier agents. Across these settings, we could see when agents successfully optimized a quantitative metric without understanding the rules underlying the system. In particular, their scientific reasoning deteriorated when they encountered systems whose rules fell outside familiar biological priors. By highlighting such failure modes, science sandboxes make the frontier of scientific capability measurable and provide a controlled setting in which to study and ultimately expand it.

Related papers

• EvoSCM: Scientific Belief Revision Through Causal Model Evolution and Experimentation [68.80901568810616]
  EvoSCM aims to equip scientific agents with explicit structural causal models that evolve as new evidence is collected.<n>We evaluate EvoSCM onPhysics, a benchmark requiring agents to uncover the hidden dynamics of noncanonical physical worlds through experimentation.
  arXiv  Detail & Related papers  (2026-09-01T16:55:56Z)
• Embodied Science: Closing the Discovery Loop with Agentic Embodied AI [73.54118756665221]
  We argue for embodied science, a paradigm that reframes scientific discovery as a closed loop tightly coupling agentic reasoning with physical execution.<n>We propose a unified Perception-Language-Action-Discovery (PLAD) framework, wherein embodied agents perceive experimental environments, reason over scientific knowledge, execute physical interventions, and internalize outcomes to drive subsequent exploration.
  arXiv  Detail & Related papers  (2026-03-20T09:20:12Z)
• SciIF: Benchmarking Scientific Instruction Following Towards Rigorous Scientific Intelligence [60.202862987441684]
  We introduce scientific instruction following: the capability to solve problems while strictly adhering to the constraints that establish scientific validity.<n>Specifically, we introduce SciIF, a multi-discipline benchmark that evaluates this capability by pairing university-level problems with a fixed catalog of constraints.<n>By measuring both solution correctness and multi-constraint adherence, SciIF enables finegrained diagnosis of compositional reasoning failures.
  arXiv  Detail & Related papers  (2026-01-08T09:45:58Z)
• Probing Scientific General Intelligence of LLMs with Scientist-Aligned Workflows [203.3527268311731]
  We present an operational SGI definition grounded in the Practical Inquiry Model (PIM)<n>We operationalize it via four scientist-aligned tasks: deep research, idea generation, dry/wet experiments, and experimental reasoning.<n>Our PIM-grounded definition, workflow-centric benchmark, and empirical insights establish a foundation for AI systems that genuinely participate in scientific discovery.
  arXiv  Detail & Related papers  (2025-12-18T12:44:36Z)
• Autonomous Agents for Scientific Discovery: Orchestrating Scientists, Language, Code, and Physics [82.55776608452017]
  Large language models (LLMs) provide a flexible and versatile framework that orchestrates interactions with human scientists, natural language, computer language and code, and physics.<n>This paper presents our view and vision of LLM-based scientific agents and their growing role in transforming the scientific discovery lifecycle.<n>We identify open research challenges and outline promising directions for building more robust, generalizable, and adaptive scientific agents.
  arXiv  Detail & Related papers  (2025-10-10T22:26:26Z)
• BoxingGym: Benchmarking Progress in Automated Experimental Design and Model Discovery [32.395366135636046]
  We introduce BoxingGym, a benchmark with 10 environments for evaluating experimental design and model discovery.<n>We compute the expected information gain (EIG), an information-theoretic quantity which measures how much an experiment reduces uncertainty about the parameters of a generative model.<n>We find that current LLMs, such as GPT-4o, struggle with both experimental design and model discovery.
  arXiv  Detail & Related papers  (2025-01-02T21:15:57Z)
• LLM and Simulation as Bilevel Optimizers: A New Paradigm to Advance Physical Scientific Discovery [141.39722070734737]
  We propose to enhance the knowledge-driven, abstract reasoning abilities of Large Language Models with the computational strength of simulations. We introduce Scientific Generative Agent (SGA), a bilevel optimization framework. We conduct experiments to demonstrate our framework's efficacy in law discovery and molecular design.
  arXiv  Detail & Related papers  (2024-05-16T03:04:10Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.