About
I currently lead a modeling team in Microsoft AI. We are working on large-scale (OpenAI and MAI) model training, with a recent focus on agentic and long-horizontal RL and synthetic data generation. We also publish some of our works, such as LoRA (opens in new tab), DeBERTa (opens in new tab), Phi (opens in new tab), Rho-1 (opens in new tab). Please check: Weizhu Chen – Google Scholar (opens in new tab) for our recent works.