UCLA Ph.D. student Harold Li in the Computer Science department received the 2023 Google Ph.D. Fellowship for his work in Natural Language Processing. The Google Ph.D. Fellowship program was created to recognize outstanding graduate students doing exceptional and innovative research in areas relevant to computer science and related fields.

Li is advised by UCLA Professor Kai – Wei Chang and a member of UCLA-NLP. He led pioneering efforts in vision-language research, including developing notable models such as VisualBERT, GLIP, and the recently introduced DesCo models. His research develops artificial intelligence systems that can perceive the visual world and communicate about it in natural languages. For example, his recent work proposes training multimodal models with easily-accessible supervision in the form of natural language. The model learns through rich descriptions of the visual world. This is akin to how humans learn about unseen objects and tasks. Training with language supervision allows the user to specify custom needs and domain knowledge at inference time; the model situates its predictions based on the specifications in the moment. This enhances the models’ ability to generalize and adapt efficiently and facilitates the ability for humans and the machines to collaborate easily.

In recognition of his contributions to vision-language research and his commitment to advancing the field of Natural Language Processing, Harold Li’s receipt of the 2023 Google Ph.D. Fellowship is a testament to his dedication in the realm of computer science and related fields.