HuggingFace Daily Papers·· 1 天前
LLM 能否通过自由探索发现隐藏的物理定律？LongChang 框架评测结果
Can LLMs Uncover Hidden Physical Laws through Free Exploration?
AI 导读
研究提出 LongChang 自动化框架，构建物理定律被改写的模拟世界，让 LLM 在无任务提示下自主操控物体、设计工具并探索规律。在 12 个 LLM 中，GPT-6-Astra 的定律发现率为 16.7%，其余 11 个均为 0%。探索记录显示模型会遗漏未调查的异常现象，且仅能部分识别被改写的物理定律。
正文 · 原文
Abstract:Scientific discovery often begins with an unexpected observation that prompts a new research question. In a world with altered physical laws, can large language models (LLMs) notice unexpected behavior, such as heavier objects falling more slowly, and discover the law through experiments without being told what to investigate? To evaluate this ability, we introduce LongChang, an automated framework that creates simulated worlds with altered physics, where one or more laws differ from ordinary physics. The evaluated LLMs act as explorers, manipulating objects and building tools to investigate their own questions without research tasks or hints about these laws. Separate LLM agents generate and review scenes, review tool designs, and evaluate reports for correctness and experimental support using the altered physical laws and exploration records. These records guide scene evolution for further exploration, allowing the full cycle to run without human intervention. Among 12 LLMs, GPT-6-Astra achieves a discovery rate of 16.7%, while the other 11 score 0%. The exploration records reveal uninvestigated anomalies and only partial identification of the altered physical laws.
| Comments: | this https URL |
| Subjects: | Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Multiagent Systems (cs.MA) |
| Cite as: | arXiv:2610.37649 [cs.CL] |
| (or arXiv:2610.37649v1 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2610.37649 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Zhen Yang [view email]
[v1]
Tue, 27 Oct 2026 17:57:00 UTC (5,338 KB)
来源：HuggingFace Daily Papers · arxiv.org