Jekyll2026-11-05T07:48:09+00:00https://refrain-wbh.github.io/feed.xmlBinghai Wang (汪冰海) is a PhD student at the Fudan University NLP Lab working on reward modeling, RLHF (reinforcement learning from human feedback), and scalable oversight for large language models. Author of RMB, WorldPM, and Secrets of RLHF.