DeepOR: A Deep Reasoning Foundation Model for Optimization Modeling
Ziyang Xiao, Yuan Jessica Wang, Xiongwei Han, Shisi Guan, Jingyan Zhu, Jingrong Xie, Lilin Xu, Han Wu
Abstract
Optimization modeling plays a critical role in supporting optimal decision-making across various domains. Previous works have demonstrated that large language models (LLMs) tailored for optimization modeling have significantly automated and simplified this process. However, these models typically employ a straightforward input-output paradigm and struggle with challenging instances. In contrast, recent advances in general-purpose reasoning LLMs (RLLMs), such as DeepSeek-R1, have shown impressive capabilities in complex domains like mathematics and coding. In this paper, we introduce DeepOR, the first RLLM specifically designed for optimization modeling. Instead of directly outputting solutions, DeepOR explicitly performs multiple intermediate reasoning steps. To adapt a base LLM into an RLLM, we begin by synthesizing long chain-of-thought (CoT) data guided by a flowchart, which is automatically generated using a self-exploration algorithm. Once the training data are prepared, we employ supervised fine-tuning on the base LLM to endow it with reasoning capabilities tailored for optimization modeling. To fully leverage the model
BibTeX
@inproceedings{aaai2026_deeporadeepreaso,
title = {DeepOR: A Deep Reasoning Foundation Model for Optimization Modeling},
author = {Ziyang Xiao and Yuan Jessica Wang and Xiongwei Han and Shisi Guan and Jingyan Zhu and Jingrong Xie and Lilin Xu and Han Wu and Wing Yin Yu and Zehua Liu and Xiaojin Fu and Gang Chen and Dongxiang Zhang},
booktitle = {AAAI 2026},
year = {2026}
}