Know When the Equation Is Wrong: Structural Uncertainty for LLM-Guided Symbolic Regression under Extrapolation
cs.LG, cs.AI
Submitted: 2026-10-30
Updated: 2026-10-30
Code: https://github.com/MostafaCham/LLM-SR-UQ
License: http://creativecommons.org/licenses/by/4.0/
The gist: Large language models (LLMs) can help discover mathematical equations from data, but most LLM-guided symbolic regression methods focus on a single best equation without quantifying uncertainty over…
Terminology
Abstract
Large language models (LLMs) can help discover mathematical equations from data, but most LLM-guided symbolic regression methods focus on a single best equation without quantifying uncertainty over alternative functional forms. This is challenging because distinct equations can fit the observed data similarly while producing substantially different extrapolations. Our experiments on the LLM-SRBench dataset show that standard conformal intervals achieve nominal in-distribution coverage but frequently fail under extrapolation. To tackle this challenge, we introduce LLM-SR-UQ, a framework that retains competing equation structures and uses their disagreement to characterize uncertainty about the underlying functional form. We extend the standard fixed-width conformal approach to quantify structural intervals, which measure disagreement among candidate equations and identify where prediction intervals should widen. Extensive experiments show our framework substantially improves coverage attainment by incorporating structural disagreement while preserving in-distribution coverage. Our approach also shows good generalizability when it is applied to other symbolic-regression methods.
Sources
- A Gentle Introduction to Conformal Prediction and Distribution-Free Uncertainty Quantification
- LLM-PDESR: Robust PDE Discovery via Subdomain Weighted Residuals and LLM-Guided Symbolic Hypothesis Generation
- Sparse Identification of Nonlinear Dynamics with Conformal Prediction
- Symbolic Regression with a Learned Concept Library
- Contemporary Symbolic Regression Methods and their Relative Performance
- Dictionaries, Not Darwin: Set-Level Selection Beats LLM Evolution in Scientific Equation Discovery
- In-Context Symbolic Regression: Leveraging Large Language Models for Function Discovery
- Influence-Guided Symbolic Regression: Scientific Discovery via LLM-Driven Equation Search with Granular Feedback
- LLM-SR: Scientific Equation Discovery via Programming with Large Language Models
- LLM-SRBench: A New Benchmark for Scientific Equation Discovery with Large Language Models
- STRIDE: A Self-Reflective Agent Framework for Reliable Automatic Equation Discovery
- Conformal Prediction Under Covariate Shift
- DrSR: LLM based Scientific Equation Discovery with Dual Reasoning from Data and Experience
- SR-Scientist: Scientific Equation Discovery With Agentic AI
- FunctionEvolve: Structure-Guided Symbolic Regression with LLMs
- Language models guide symbolic equation discovery by controlling search
- Think like a Scientist: Physics-guided LLM Agent for Equation Discovery
- A-SR: Self-Evolving Agentic LLMs for Symbolic Regression via Hierarchical Coordination
Related papers
- A Physics-Regulated Neural Framework for Learning 3D Grain Growth Dynamics
- FlexMoE: Learning One-for-All Multi-Budget Expert Pruning for Flexible MoE Deployment
- OptiMind: Improving Small Language Models for Optimization Modeling through Formulation Guidance and Multi-Turn Learning
- Staying in the Sweet Spot: Boost Reasoning Evolution via Responsive Rollout Difficulty Manipulation
- Examining the Impact of Data and Training Methods on Emergent and Subliminal Misalignment
- Rethinking Black-Box On-Policy Distillation: A Rubric-based Analysis