David Yao

DiFFPO: Training Diffusion LLMs to Reason Fast and Furious via Reinforcement Learning

Understanding Sampler Stochasticity in Training Diffusion Models for RLHF

MallowsPO: Fine-Tune Your LLM with Preference Dispersions

Score as Action: Fine Tuning Diffusion Generative Models by Continuous-time Reinforcement Learning

RainbowPO: A Unified Framework for Combining Improvements in Preference Optimization