Hao Liu
I'm a research scientist at Google DeepMind.
Previously, I was a PhD student in EECS at Berkeley, advised by Professor Pieter Abbeel, and spent two years part-time at Google Brain.
I'm an incoming Assistant Professor at Carnegie Mellon University starting in Fall 2025.
I work on solving general superintelligence. My interests are designing scalable neural architectures and training objectives.
News:
ElasticTok: adapts computation based on available information in images/videos (paper'24,code).
Large World Models(LWM) for modeling text and video of millions tokens length (code).
Agentic Transformer for learning decision-making from trial-and-error sequences at scale (paper'23).
Nearinfinite context with BlockwiseTransformer and RingAttention (paper'24,paper'23,code).
Open language models: Koala, OpenLLaMa (blog'23,models).
Reinforcement learning from unsupervised exploration (paper'22,paper'22,paper'21).