TxT360 WORCS: an Open Recipe and Framework for Language Model Pretraining Data (Invited talk: Hector Zhengzhong Liu)

Hector Liu is the Director of the Institute of Foundation Models at the Silicon Valley Lab, where he leads research on foundation models and AI. He has spearheaded projects such as LLM360, an initiative for fully open foundation models; K2, the most performant open model; and Jais, the leading Arabic language model. Previously, he served as the Head of Engineering at Petuum Inc. Hector earned his PhD in Natural Language Processing and Computational Linguistics from Carnegie Mellon University, advised by Professors Teruko Mitamura and Eduard Hovy. He also collaborated closely with Professor Eric Xing on the CASL project. His introduction to NLP began at the Hong Kong Polytechnic University under the guidance of Professor Qin Lu.