STRIDE: A Self-Reflective Agent Framework for Reliable Automatic Equation Discovery

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Su, Jiarui, Tu, Songjun, Sun, Bei, Liang, Xiaojun
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917506595160064
author Su, Jiarui
Tu, Songjun
Sun, Bei
Liang, Xiaojun
author_facet Su, Jiarui
Tu, Songjun
Sun, Bei
Liang, Xiaojun
contents LLM-based equation discovery offers a promising route to recovering symbolic laws from data, but many systems still rely on generation-centered loops that propose candidates, fit parameters, score results, and reuse selected examples. Such loops can misjudge useful skeletons under unreliable fitting, discard near-correct equations that require repair, and accumulate redundant memories that provide limited guidance. We propose STRIDE, a self-reflective agent framework that improves reliability by coordinating data-aware generation, mixed-fitting evaluation, critic--executor repair, and diversity-preserving semantic memory. By turning fitted scores and candidate behavior into shared feedback, STRIDE enables equations to be proposed, assessed, refined, and reused within a closed-loop discovery process. Experiments on representative symbolic-regression benchmarks and LSR-Synth suites show that STRIDE improves accuracy, OOD robustness, and structural recovery across multiple LLM backbones, with ablations and analyses confirming the contribution of its core components.
format Preprint
id arxiv_https___arxiv_org_abs_2605_17790
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle STRIDE: A Self-Reflective Agent Framework for Reliable Automatic Equation Discovery
Su, Jiarui
Tu, Songjun
Sun, Bei
Liang, Xiaojun
Artificial Intelligence
68T05, 68T20, 68T42, 68W30
I.2.6; I.2.8; I.1.2
LLM-based equation discovery offers a promising route to recovering symbolic laws from data, but many systems still rely on generation-centered loops that propose candidates, fit parameters, score results, and reuse selected examples. Such loops can misjudge useful skeletons under unreliable fitting, discard near-correct equations that require repair, and accumulate redundant memories that provide limited guidance. We propose STRIDE, a self-reflective agent framework that improves reliability by coordinating data-aware generation, mixed-fitting evaluation, critic--executor repair, and diversity-preserving semantic memory. By turning fitted scores and candidate behavior into shared feedback, STRIDE enables equations to be proposed, assessed, refined, and reused within a closed-loop discovery process. Experiments on representative symbolic-regression benchmarks and LSR-Synth suites show that STRIDE improves accuracy, OOD robustness, and structural recovery across multiple LLM backbones, with ablations and analyses confirming the contribution of its core components.
title STRIDE: A Self-Reflective Agent Framework for Reliable Automatic Equation Discovery
topic Artificial Intelligence
68T05, 68T20, 68T42, 68W30
I.2.6; I.2.8; I.1.2
url https://arxiv.org/abs/2605.17790