A Sampler’s Blog
Welcome to Wei Deng’s homepage. My research focuses on fast reasonings in (diffusion) LLMs. I also study generative methods for deep hedging. Here I share ideas I find insightful on my own works.
Contact me at: firstnamelastname056@gmail.com. Also looking for tennis enthusiast in NJ.
🔥 Welcome! Self-Correcting Discrete Diffusion — Faster parallel decoding w/o remasks.
🔥 Welcome! Group Diffusion Policy Optimization — dLLM reasoning in math and coding. ICLR’26 [Link]
🔥 Welcome! Ultra-Fast Language Generation — 60× faster than GPT-2, same quality. ICLR’26 [X]
La pensée n’est qu’un écliar au milieu d’une longue nuit. Mais c’est cet éclair qui est tout. - Poincaré