PhD student, Northeastern University
1 paper at NeurIPS 2025
We propose the Multi-Reward Optimization (MRO) approach, which enhances token correlation during the denoising process in diffusion language models, improving reasoning performance and sampling efficiency.
