Skip to content

SymbolicLM: Training Language Models as Symbolic Regressors

Sep 2026 · arXiv.org · Vol abs/2609.33594 · 0 citations · 21 references
Computer Science

TL;DR

Experiments on multiple symbolic regression benchmarks show that SymbolicLM substantially improves structural recovery while maintaining competitive numerical fitting performance, demonstrating that symbolic regression can be explicitly learned as an intrinsic capability of LLMs.

Abstract

Large Language Models (LLMs) have shown promising capabilities in scientific reasoning, yet scientific discovery ultimately requires deriving precise laws directly from observational data, known as Symbolic Regression (SR). This poses a challenge for LLMs due to the gap between probabilistic text generation and the exact structural requirements of SR. Existing approaches rely on complex external scaffolds, which are computationally expensive and separate symbolic reasoning from the model itself. To address this limitation, we propose to directly equip LLMs with symbolic regression capabilities through dedicated numerical-symbolic and physical supervision. We introduce PhysSymbArena, a large-scale benchmark containing over 160,000 equations and 1.8B tokens of numerical-symbolic data with physical descriptions, enabling systematic training and evaluation. Based on PhysSymbArena, we develop SymbolicLM, which enhances the symbolic regression ability of LLMs through mathematical and physical supervision. During inference, we further introduce SymbolicSGA, a refinement framework that leverages quantitative feedback to iteratively improve generated equations. Experiments on multiple symbolic regression benchmarks show that SymbolicLM substantially improves structural recovery while maintaining competitive numerical fitting performance. These results demonstrate that symbolic regression can be explicitly learned as an intrinsic capability of LLMs.

View source

Similar papers

#large language models Open access Sep 2026

Iterated Agent for Symbolic Regression

This paper introduces IdeaSearchFitter, a framework that employs Large Language Models (LLMs) as semantic operators within an evolutionary search that achieves competitive, noise-robust performance on the Feynman Symbolic Regression Database (FSReD), outperforming several strong baselines.

Zhuo-Yang Song, Ze-Yu Cai, Shu-Tao Zhang et al. · 0 citations
#artificial intelligence Preprint Sep 2026

SRHarness: A Harness for Agentic Symbolic Regression

SRHarness is introduced, a domain-specific harness for agentic symbolic regression built around three mechanisms: composable scientific actions that provide a common interface over raw, transformed, and candidate-derived quantities; persistent scientific state that retains evaluated hypotheses and exposes compact model...

Zihan Yu, Shi-Xuan Zhou, Hao Huang et al. · 0 citations

SNIP++: Fine-Grained Symbolic-Numerical Alignment for Symbolic Regression

A compositional alignment method is introduced that closes much of the modality gap between symbolic and numerical embeddings, reliably distinguish the effects of local edits that the original alignment cannot, and transfer to external SR corpora.

Benjamin Léger, Shubham Gupta, S. Mammeri et al. · 0 citations
#machine learning Preprint Sep 2026

SMILE: Bridging Continuous Optimization and Discrete Symbolic Recovery

Symbolic regression (SR) discovers closed-form mathematical expressions from data, offering interpretability beyond black-box models. Existing methods suffer from slow convergence in combinatorial search spaces and lack mechanisms to exploit compositional structure in the data. We introduce SMILE (Sine, Multiplication,...

Mansooreh Montazerin, Antonio Ortega, Ajitesh Srivastava · 0 citations
#artificial intelligence Preprint Oct 2026

RISR: Residual-Informed Scientific Equation Discovery with Large Language Models

Symbolic regression combines structural search with numerical fitting, but aggregate fit scores do not describe how the remaining error varies across inputs. We introduce RISR, a residual-informed method that uses these error patterns to guide formula discovery and learn which corrections are worth fitting. A residual...

Hao-Bo Li, Wen-Shuo Zhang, Wen-Xiao Zhao et al. · 0 citations

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.