Xijun Li

Offline Reinforcement Learning via Action-Space Pseudo-Labeling

Differentiable Integer Linear Programming

$STRCMP$: Integrating Graph Structural Priors with Language Models for Combinatorial Optimization

Dynamic Configuration for Cutting Plane Separators via Reinforcement Learning on Incremental Graph

ASDSV: Multimodal Generation Made Efficient with Approximate Speculative Diffusion and Speculative Verification

Apollo-MILP: An Alternating Prediction-Correction Neural Solving Framework for Mixed-Integer Linear Programming

LLM4Solver: Large Language Model for Efficient Algorithm Design of Combinatorial Optimization Solver