Hao Zhu

he/him/他

朱昊

Postdoctoral

Cooperation, Language, CogSci, AI, Robotics.

zhuhao@stanford.edu

I am a Postdoctoral Scholar at the Stanford NLP Group working with Diyi Yang. Previously, I obtained my PhD degree from the Language Technologies Institute, Carnegie Mellon University advised by Graham Neubig and Yonatan Bisk. My thesis work was on investigating the challenges, building realistic environments, and developing methods for creating socially intelligent artificial agents.

I take an interactionist approach to understand how language is acquired and used in the context of social and embodied environments. Previously I studied Theory-of-Mind modeling and proactive exploration's role in language learning. Now I am focusing on building realistic open-ended social and embodied environments. I have released 3 of these: Sotopia, WebArena, and Excalibur. Please also check out our newest SOTOPIA-π on training social agents in the Sotopia environment.

My ultimate dream is to unravel the mysteries of human intelligence. Inspired by Feynman's famous quote, "What I cannot create, I do not understand," I dive into research in ML, NLP, and Robotics, not only taking inspiration from but also contributing to a better understanding of humans themselves.