Wei Chen

Benefits and Pitfalls of Reinforcement Learning for Language Model Planning: A Theoretical Perspective

Combinatorial Rising Bandits

Understanding and Enhancing the Planning Capability of Language Models via Multi-Token Prediction

Multi-play Combinatorial Semi-Bandit Problem

Hybrid Combinatorial Multi-armed Bandits with Probabilistically Triggered Arms

PaT: Planning-after-Trial for Efficient Code Generation

Regret Analysis of Hybrid Linear Bandits with Biased Offline Data

Planner and Executor: Collaboration between Discrete Diffusion And Auto-regressive Models in Reasoning

Mechanism Design for LLM Fine-tuning with Multiple Reward Models

Structural-Entropy-Based Sample Selection for Efficient and Effective Learning

Offline Learning for Combinatorial Multi-armed Bandits

Cost-Efficient Multi-Fidelity Alignment for LLMs