✳flâneur — a map of the web's best reading

Percy Liang

cs.stanford.edu · 1,022 words · saved by 1 readers

If you are a Stanford undergraduate or masters student interested in joining my research group, please apply here. I publish mostly in machine learning (ICML, NeurIPS) and natural language processing (ACL, NAACL, EMNLP) conferences. I am drawn to simple things, want to understand things deeply, and like to build useful systems. These days, I am fascinated by foundation models (e.g., GPT-4, PaLM, DALL-E), models trained on broad data using self-supervision at immense scale, which can then be adapted to myriad downstream tasks. Foundation models represent a paradigm shift in how AI systems are built and how people might interact with AI. How can we better understand how they work and make them more efficient, modular, and robust? I am also a strong proponent of efficient and reproducible research. We have been developing CodaLab Worksheets, a platform that allows researchers to run and manage their experiments by maintaining the full provenance of an experiment from raw data to final res

Percy Liang Percy Liang Professor of Computer Science (and courtesy in Statistics) Gates 366 [Papers] [Marin] Research I am drawn to simple things, want to understand things deeply, and like to build useful systems. I am interested in fundamental questions around learning and intelligence. How well can we learn in the data-limited, infinite compute regime? How do we model interestingness and curiosity? How can we push the frontier of human knowledge? I lead Marin to build models openly. Here is a talk I gave about it. We practice a radical form of openness called open development, which goes b

Explore this link on the map →

related reading