Explore, Then Commit: Measurement-Efficient Scientific Law Discovery with Language Models

cs.AI, cs.LG, physics.comp-ph

Submitted: 2026-10-06

Updated: 2026-10-06

Comments: 19 pages, including Supplementary Material S1; code and data included as ancillary files. Preprint

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Sources

• NewtonBench: Benchmarking Generalizable Scientific Law Discovery in LLM Agents
• Interpretable Machine Learning for Science with PySR and SymbolicRegression.jl
• Contemporary Symbolic Regression Methods and their Relative Performance
• LLM-SR: Scientific Equation Discovery via Programming with Large Language Models
• BoxingGym: Benchmarking Progress in Automated Experimental Design and Model Discovery
• Bayesian Active Learning for Classification and Preference Learning
• BED-LLM: Intelligent Information Gathering with LLMs and Bayesian Experimental Design
• PAL: Program-aided Language Models
• Program of Thoughts Prompting: Disentangling Computation from Reasoning for Numerical Reasoning Tasks

Related papers

• MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
• Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
• Agents' Last Exam
• The Clinician's Veto: Navigating Trust, Liability, and Uncertainty in Autonomous AI Prescribing
• MindHelper: Closed-Loop Embodied Mental-State Reasoning for Precision Intervention
• Incumbent Advantage: Brand Bias and Cognitive Manipulation Dynamics in LLM Recommendation Systems