Title: Towards Foundation Models for Solving Vehicle Routing Problems

Abstract: The Vehicle Routing Problem (VRP) is a cornerstone of combinatorial optimization in operations research and is traditionally tackled using heuristics with hand-crafted rules. In recent years, there has been growing interest in leveraging deep learning, including large language models (LLMs), to automatically discover and learn such heuristics. In this talk, I will first provide an overview of neural approaches for solving individual VRPs. I will then delve into recent advances toward building foundation models capable of generalizing across a wide range of VRP variants. Finally, I will highlight the key challenges in this emerging field and share my perspectives on future directions.

Bio: Dr. Zhiguang Cao is an Assistant Professor at the School of Computing and Information Systems, Singapore Management University (SMU). He received his Ph.D. from Nanyang Technological University, Singapore in 2017. His research centers on Learning to Optimize (L2Opt), where deep learning techniques (including LLM) are applied to solve classical combinatorial optimization problems such as the vehicle routing, job-shop scheduling, and bin packing. Dr. Cao has published 28 papers at ICML, NeurIPS, and ICLR, where he also serves regularly as an Area Chair.