DAOpt: Modeling and Evaluation of Data-Driven Optimization under Uncertainty with LLMs

2025/09/24 by Zhu, WenZhuo, Zheng Cui, Wenhan Lu +5
Computer Science · #Artificial Intelligence (cs.AI) #Constraint Satisfaction and Optimization #FOS: Computer and information sciences #Machine Learning (cs.LG) #Multimodal Machine Learning Applications #Topic Modeling

paper · pdf · doi:10.48550/arxiv.2511.11576

openalex publication_date 2025/09/24 · openalex created_date 2025/11/19 · openalex updated_date 2026/07/28

Abstract

Recent advances in large language models (LLMs) have accelerated research on automated optimization modeling. While real-world decision-making is inherently uncertain, most existing work has focused on deterministic optimization with known parameters, leaving the application of LLMs in uncertain settings largely unexplored. To that end, we propose the DAOpt framework including a new dataset OptU, a multi-agent decision-making module, and a simulation environment for evaluating LLMs with a focus on out-of-sample feasibility and robustness. Additionally, we enhance LLMs' modeling capabilities by incorporating few-shot learning with domain knowledge from stochastic and robust optimization.

Citations

• DeepSeek-R1 incentivizes reasoning in LLMs through reinforcement learning
• LLMOPT: Learning to Define and Solve General Optimization Problems from Scratch
• OptiBench Meets ReSocratic: Measure and Improve LLMs for Optimization Modeling
• DeepSeek-Coder-V2: Breaking the Barrier of Closed-Source Models in Code Intelligence
• ORLM: A Customizable Framework in Training Large Models for Automated Optimization Modeling
• OptiMUS: Scalable Optimization Modeling with (MI)LP Solvers and Large Language Models
• Progressive-Hint Prompting Improves Reasoning in Large Language Models
• Reflexion: Language Agents with Verbal Reinforcement Learning
• GPT-4 Technical Report
• BNAI, NO-TOKEN, and MIND-UNITY: Pillars of a Systemic Revolution in Artificial Intelligence
• Language Models are Few-Shot Learners
• Data-driven Distributionally Robust Optimization Using the Wasserstein\n Metric: Performance Guarantees and Tractable Reformulations
• Data-driven distributionally robust optimization using the Wasserstein metric: performance guarantees and tractable reformulations
• The Price of Robustness

Related

• A Survey of Optimization Modeling Meets LLMs: Progress and Future Directions
• Step-Opt: Boosting Optimization Modeling in LLMs through Iterative Data Synthesis and Structured Validation
• When Large Language Model Meets Optimization
• GPTOpt: Towards Efficient LLM-Based Black-Box Optimization
• ORLM: A Customizable Framework in Training Large Models for Automated Optimization Modeling