Language models guide symbolic equation discovery by controlling search

arXiv cs.AI Papers

Summary

This paper introduces LLM-PySR, a method where language models guide symbolic equation discovery by controlling search parameters while using numerical symbolic regression for fitting. The approach achieves strong balance of accuracy and complexity across benchmark tasks.

arXiv:2607.04156v1 Announce Type: new Abstract: Scientific equation discovery must combine broad domain priors with strict numerical testing. Symbolic regression supplies numerical grounding but faces a combinatorial search space, whereas many language-model systems ask the model to propose or select formulas directly. We test a different division of labour. We compare role specifications in which the language model acts as equation author, candidate decider or search controller, alongside end-to-end language-model and purely numerical baselines. In the controller setting we propose here, implemented as LLM-PySR, language models specify variables, operators, transformations and search depth; symbolic regression enumerates and fits expressions; and deterministic metrics govern retention. Across 74 AI-Feynman equations and seven complex formula-recovery tasks, search control achieved the strongest observed balance of accuracy, complexity, stability and cost. On an independent battery dataset, LLM-PySR identified a compact piecewise-linear relation between early voltage-curve displacement and cycle life. The results suggest that language models should shape hypothesis exploration rather than decide which equations survive.
Original Article
View Cached Full Text

Cached at: 07/07/26, 04:36 AM

# Language models guide symbolic equation discovery by controlling search
Source: [https://arxiv.org/abs/2607.04156](https://arxiv.org/abs/2607.04156)
[View PDF](https://arxiv.org/pdf/2607.04156)

> Abstract:Scientific equation discovery must combine broad domain priors with strict numerical testing\. Symbolic regression supplies numerical grounding but faces a combinatorial search space, whereas many language\-model systems ask the model to propose or select formulas directly\. We test a different division of labour\. We compare role specifications in which the language model acts as equation author, candidate decider or search controller, alongside end\-to\-end language\-model and purely numerical baselines\. In the controller setting we propose here, implemented as LLM\-PySR, language models specify variables, operators, transformations and search depth; symbolic regression enumerates and fits expressions; and deterministic metrics govern retention\. Across 74 AI\-Feynman equations and seven complex formula\-recovery tasks, search control achieved the strongest observed balance of accuracy, complexity, stability and cost\. On an independent battery dataset, LLM\-PySR identified a compact piecewise\-linear relation between early voltage\-curve displacement and cycle life\. The results suggest that language models should shape hypothesis exploration rather than decide which equations survive\.

## Submission history

From: Zikai Xie \[[view email](https://arxiv.org/show-email/7b9d36f5/2607.04156)\] **\[v1\]**Sun, 5 Jul 2026 07:50:08 UTC \(2,196 KB\)

Similar Articles

Learning to reason with LLMs

OpenAI Blog

OpenAI publishes an article exploring reasoning techniques with LLMs through cipher-decoding examples, demonstrating step-by-step problem-solving approaches and pattern recognition in language models.

Model Unlearning Objectives Vary for Distinct Language Functions

arXiv cs.CL

The paper argues that unlearning in LLMs should be goal-dependent, proposing a cosine-based meta-learned variant of RMU for dangerous knowledge and a multi-layer objective with probe directions for toxicity, achieving strong results across four 7-8B models.