L2G-SR is a framework that enhances large language model (LLM)-based symbolic regression by incorporating a learned structural verifier. The verifier uses tree edit distance (TED) to assess structural similarity between candidate equations and the target, complementing numerical error during search. Experiments on LLM-SRBench show that L2G-SR achieves state‑of‑the‑art performance in numerical fit, equation‑level accuracy, and TED across various LLM backbones and domains.
arXiv:2609.36187v1 Announce Type: new Abstract: Symbolic Regression (SR) is a data-driven method for scientific discovery which searches for interpretable analytical relationships within data. Recent...
arXiv:2605. 23272v2 Announce Type: replace-cross Abstract: Symbolic Regression (SR) plays a central role in scientific knowledge discovery by distilling mathematical equations from observational data.
arXiv:2606. 07704v1 Announce Type: cross Abstract: Symbolic regression aims to uncover explicit scientific laws from data.
arXiv:2606. 04360v1 Announce Type: cross Abstract: Symbolic regression (SR) discovers compact mathematical expressions from data, yet recent LLM-based evolutionary methods remain sample-inefficient because they rely mainly on scalar feedback such as MSE.
The paper introduces RISR, a residual‑informed approach for symbolic regression that leverages error patterns to guide formula discovery. A residual encoder transforms aligned inputs, targets, predictions, and residuals into tokens that condition a language model to propose candidate equations, while a dual‑view relational encoder predicts the utility of corrections. Evaluated on the LLM‑SRBench, RISR achieves higher in‑distribution and out‑of‑distribution accuracy than baseline methods, demonstrating that incorporating residual information can improve numerical equation recovery.
The paper presents a probabilistic symbolic regression framework that models mathematical expressions as ensembles of symbolic trees, using a regularizing prior to control complexity and an Occam’s window-based posterior to capture uncertainty across plausible models. It provides theoretical guarantees on posterior concentration, including near‑parametric rates when an exact finite formula exists and oracle results under misspecification. Empirical results show the method outperforms state‑of‑the‑art competitors in predictive accuracy, symbolic complexity, and structural recovery on benchmark scientific equations and a materials discovery task.
InsightSR is a new framework that integrates Large Language Models (LLMs) with the PySR genetic programming engine to refine symbolic regression search spaces. It employs two LLM-guided pathways: a Semantic Seed Pathway that generates dimensionally consistent functional skeletons, and a Structural Feature Pathway that suggests nonlinear feature transformations. Over successive iterations, these pathways expand the input space and shift the search toward shallow, semantically informed trees, with a feedback loop that evaluates and refines candidate features. The method achieves a 95% exact recovery rate on the Feynman benchmark and 80.18% accuracy on the LLM-SRBench LSR-Transform task, outperforming existing genetic programming and neural-symbolic approaches while preserving strong out-of-distribution generalization.
DIGEST is a new method for symbolic regression that combines a lightweight LLM with Genetic Programming. It uses a pretrained decoder block as a symbolic policy to seed GP populations, providing flexible generation of candidate expressions. The approach also feeds high‑quality GP discoveries back into the LLM to refine future generations, leading to improved recovery rates and faster search times on benchmark datasets.
arXiv:2608. 16876v1 Announce Type: cross Abstract: We introduce Automatic Symbolic Regression (AutoSR), a fully automated system that instantiates Research-Space Symbolic Regression by searching persistent scientific investigations rather than isolated equations.
arXiv:2606. 09276v1 Announce Type: new Abstract: Equation discovery aims to automate the discovery of scientific models in the form of mathematical equations from data.
arXiv:2606. 07915v1 Announce Type: new Abstract: Neural symbolic regression models improve inference efficiency by shifting structural search to pretraining, but their one-pass autoregressive decoding is prone to error accumulation, which may lead to generating structurally incorrect expressions, especially in complex expression generation scenarios.
Symbolic regression aims to discover closed-form equations from data, but existing LLM-guided methods often rely on a unified proposal loop that compresses heterogeneous search failures into a scalar score and a single prompt. We propose A-SR, a self-evolving agentic framework that shifts the control unit from expression edits to role-conditioned evidence views.