Prof. Dongdong Ge
Shanghai Jiao Tong University
Talk: From ORLM to Decision Systems: Self-Evolving LLMs for Optimization and Beyond
Abstract: Optimization modeling is central to operations research, yet translating real-world problems into correct mathematical formulations remains a major bottleneck. Recent advances in large language models (LLMs) offer a new pathway, but challenges in reliability, scalability, and continual improvement persist. In this talk, we present a unified framework for self-evolving LLM-based decision systems. We develop training-based approaches (ORLM, Solver-Informed RL, StepORLM), extend modeling to dynamic programming, and analyze intrinsic limits via OPT-Engine. Moving beyond static models, we introduce agentic systems with solver-grounded validation, decentralized debate, and memory, enabling reliable and reusable decision-making. These results suggest a new paradigm in which LLMs evolve from generators into verifiable, adaptive decision systems for operations research and beyond.
Biography: Dongdong Ge is a professor at the Antai College of Economics and Management of Shanghai Jiao Tong University and the dean of the Institute for Intelligent Computing (SJTU-IIC). He obtained his PhD from the Department of Management Science and Engineering at Stanford University in 2009. His main research interests are in the theory, computation, and application of large-scale optimization problems. He publishes papers in journals and conferences across multiple fields such as MS/OR/TCS/AI, including OR, MOR, MP, POM, FOCS, SODA, EC, ICML, NeurIPS, etc. He is also a co-founder and chief scientist of Cardinal Operations(shanshu.ai) and has participated in numerous optimization projects with companies and organizations worldwide, such as Boeing, Google, Huawei, JD.com, SF Express, DiDi, State Grid Corporation of China/Southern Power Grid, and China Southern Airlines. Currently, he serves as the project leader for China's first open-source mathematical programming software LEAVES and the first commercial mathematical programming software COPT project.
