analiza aiqbench
FunctionEvolve addresses the "structure-blind" limitation of current LLM-driven symbolic regression by integrating expression trees to enforce valid symbolic representations and local mutations. This architectural shift enables reliable recovery of exact scientific laws from data, significantly outperforming baselines on the LLM-SRBench benchmark while avoiding brittle coefficient fitting. The work also highlights critical benchmark integrity issues, such as collinearity in material science subsets, which can create identifiability problems for existing evaluation metrics.
FunctionEvolve: Structure-Guided Symbolic Regression with LLMs
arXiv:2606.07704v1 Announce Type: new Abstract: Symbolic regression aims to uncover explicit scientific laws from data. Recent methods use LLMs to guide mutation from background text, which is more directed than random genetic programming. However, exact symbolic recovery requires both semantic guidance and explicit structure, so that domain-informed search are carried out through valid symbolic representation. Current LLM-driven systems remain structure-blind: they select among opaque candidates, lack explicit mechanisms for local mutation, and rely on brittle coefficient fitting that can undervalue correct skeletons. We propose FunctionEvolve, an evolutionary framework using expression trees to organize the whole search: structural summaries promote diverse parent selection, local tree edits preserve useful subexpressions, and structure-aware fitting decomposes, constrains, and simplifies coefficients for more reliable scoring. It uses only elementary function families, without additional domain-specific rules limiting generalization. On the 129-task synthetic subset of LLM-SRBench, FunctionEvolve with \emph{Claude Opus 4.6} recovers 107 exact forms, reaching 82.9% SA@50, 4.5x above same-backbone baselines, and 55.8% SA@1, 3.6x above the strongest previously published top-1 result. Ablations show that structure-visible search is central to reliable recovery, with LLM-guided refinements and structure-aware coefficient optimization serving as essential proposal and scoring mechanisms. We also audit the benchmark and show that collinearity in its materials-science subset creates identifiability issues.
Abstract:Symbolic regression aims to uncover explicit scientific laws from data. Recent methods use LLMs to guide mutation from background text, which is more directed than random genetic programming. However, exact symbolic recovery requires both semantic guidance and explicit structure, so that domain-informed search are carried out through valid symbolic representation. Current LLM-driven systems remain structure-blind: they select among opaque candidates, lack explicit mechanisms for local mutation, and rely on brittle coefficient fitting that can undervalue correct skeletons. We propose FunctionEvolve, an evolutionary framework using expression trees to organize the whole search: structural summaries promote diverse parent selection, local tree edits preserve useful subexpressions, and structure-aware fitting decomposes, constrains, and simplifies coefficients for more reliable scoring. It uses only elementary function families, without additional domain-specific rules limiting generalization. On the 129-task synthetic subset of LLM-SRBench, FunctionEvolve with \emph{Claude Opus 4.6} recovers 107 exact forms, reaching 82.9% SA@50, 4.5x above same-backbone baselines, and 55.8% SA@1, 3.6x above the strongest previously published top-1 result. Ablations show that structure-visible search is central to reliable recovery, with LLM-guided refinements and structure-aware coefficient optimization serving as essential proposal and scoring mechanisms. We also audit the benchmark and show that collinearity in its materials-science subset creates identifiability issues.
Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
arXiv:2606.07704 [cs.LG]
(or arXiv:2606.07704v1 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2606.07704 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 298 cuvinte
Subiecte conexe
DASyR-LLM: Domain-Aware Symbolic Regression with LLMs for Kinetic Model Discovery
arXiv · 6 aug. 2026
InsightSR: Refining Symbolic Regression Search Spaces via Parallel Semantic and Structural LLM Guidance
arXiv · 27 aug. 2026
Deliberate Evolution: Agentic Reasoning for Sample-Efficient Symbolic Regression with LLMs
arXiv · 5 iun. 2026
Learning to Design SR Solvers: Harness Optimization for Symbolic Regression
arXiv · 14 oct. 2026