collaborators

8 papers

cs.LG2026

Asymptotically Optimal Regret for Reinforcement Learning without Horizon Dependence

Runlong Zhou, Zihan Zhang, Maryam Fazel +1

We study horizon-free regret minimization for finite-horizon time-homogeneous tabular Markov decision processes with states, actions, horizon , and per-trajectory total…

cs.LG2026

Understanding the Performance Gap in Preference Learning: A Dichotomy of RLHF and DPO

Ruizhe Shi, Minhak Song, Runlong Zhou +3

We present a fine-grained theoretical analysis of the performance gap between two-stage reinforcement learning from human feedback~(RLHF) and direct preference optimization~(DPO).…

cs.LG2026

Unregularized Linear Convergence in Zero-Sum Game from Preference Feedback

Shulun Chen, Runlong Zhou, Zihan Zhang +2

Aligning large language models (LLMs) with human preferences has proven effective for enhancing model capabilities, yet standard preference modeling using the Bradley-Terry model a…

cs.LG2025

Extragradient Preference Optimization (EGPO): Beyond Last-Iterate Convergence for Nash Learning from Human Feedback

Runlong Zhou, Maryam Fazel, Simon S. Du

Reinforcement learning from human feedback (RLHF) has become essential for improving language model capabilities, but traditional approaches rely on the assumption that human prefe…

cs.LG2025

Sharp Gap-Dependent Variance-Aware Regret Bounds for Tabular MDPs

Shulun Chen, Runlong Zhou, Zihan Zhang +2

We consider the gap-dependent regret bounds for episodic MDPs. We show that the Monotonic Value Propagation (MVP) algorithm achieves a variance-aware gap-dependent regret bound of…

cs.LG2025

The Crucial Role of Samplers in Online Direct Preference Optimization

Ruizhe Shi, Runlong Zhou, Simon S. Du

Direct Preference Optimization (DPO) has emerged as a stable, scalable, and efficient solution for language model alignment. Despite its empirical success, the optimization propert…