Opt-Verifier: Unleashing the Power of LLMs for Optimization Modeling via Dual-Side Verification

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Liu, Haoyang, Wang, Jie, Niu, Boxuan, Han, Xiongwei, Xu, Yian, Ye, Mingxuan, Geng, Zijie, Zhu, Fangzhou, Zhong, Tao, Yuan, Mingxuan, Hao, Jianye
Formato: Preprint
Publicado: 2026
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866911726926036992
author Liu, Haoyang
Wang, Jie
Niu, Boxuan
Han, Xiongwei
Xu, Yian
Ye, Mingxuan
Geng, Zijie
Zhu, Fangzhou
Zhong, Tao
Yuan, Mingxuan
Hao, Jianye
author_facet Liu, Haoyang
Wang, Jie
Niu, Boxuan
Han, Xiongwei
Xu, Yian
Ye, Mingxuan
Geng, Zijie
Zhu, Fangzhou
Zhong, Tao
Yuan, Mingxuan
Hao, Jianye
contents Building mathematical optimization models is critical in operations research (OR), while it requires substantial human expertise. Recent advancements have utilized large language models (LLMs) to automate this modeling process. However, existing works often struggle to verify the correctness of the generated optimization models, without checking the rationality of the constraints and variables or the validity of solutions to the generated models. This hampers the subsequent verification and correction steps, and thus it severely hurts the modeling accuracy. To address this challenge, we propose a novel LLM-based framework with Dual-side Verification (Opt-Verifier) from both structure and solution perspectives, thereby improving the modeling accuracy. The structure-side verification ensures that the modeling structure of the generated optimization models aligns with the original problem description, accurately capturing the problem's constraints and requirements. Meanwhile, the solution-side verification interprets and evaluates the solutions' validity, confirming that the optimization models are logically and mathematically sound. Experiments on popular benchmarks demonstrate that our approach achieves over 20\% improvement in accuracy.
format Preprint
id arxiv_https___arxiv_org_abs_2605_29556
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Opt-Verifier: Unleashing the Power of LLMs for Optimization Modeling via Dual-Side Verification
Liu, Haoyang
Wang, Jie
Niu, Boxuan
Han, Xiongwei
Xu, Yian
Ye, Mingxuan
Geng, Zijie
Zhu, Fangzhou
Zhong, Tao
Yuan, Mingxuan
Hao, Jianye
Artificial Intelligence
Building mathematical optimization models is critical in operations research (OR), while it requires substantial human expertise. Recent advancements have utilized large language models (LLMs) to automate this modeling process. However, existing works often struggle to verify the correctness of the generated optimization models, without checking the rationality of the constraints and variables or the validity of solutions to the generated models. This hampers the subsequent verification and correction steps, and thus it severely hurts the modeling accuracy. To address this challenge, we propose a novel LLM-based framework with Dual-side Verification (Opt-Verifier) from both structure and solution perspectives, thereby improving the modeling accuracy. The structure-side verification ensures that the modeling structure of the generated optimization models aligns with the original problem description, accurately capturing the problem's constraints and requirements. Meanwhile, the solution-side verification interprets and evaluates the solutions' validity, confirming that the optimization models are logically and mathematically sound. Experiments on popular benchmarks demonstrate that our approach achieves over 20\% improvement in accuracy.
title Opt-Verifier: Unleashing the Power of LLMs for Optimization Modeling via Dual-Side Verification
topic Artificial Intelligence
url https://arxiv.org/abs/2605.29556