A Deeper Understanding of Large Language Models

Peter West is a Canada CIFAR AI Chair at Amii and an assistant professor at the University of British Columbia. His research focuses on understanding the capabilities and limits of generative AI systems, particularly large language models (LLMs).

Peter is especially interested in abilities that go beyond the purely technical, e.g., how do these systems navigate creativity and social intelligence? His work also aims to understand how dominant approaches may be limiting models, for instance, exploring the limitations caused by AI alignment techniques such as reinforcement learning from human feedback. Overall, Peter seeks to understand LLMs as complex scientific artifacts that warrant exploration, rather than simply the starting point for engineered systems.

Before joining UBC, Peter was a postdoctoral researcher at the Stanford Institute for Human Centered AI and the linguistics department. Prior to that, he completed his PhD at the Paul G Allen School at the University of Washington, and his BSc at the University of British Columbia. His work has been recognized as best, outstanding, and spotlight papers in NLP and AI conferences.

Awards and Highlights

• Member of the AIM-SI cluster, UBC (2025 - current)

• Outstanding Paper Award, Conference on Empirical Methods in Natural Language Processing (2023)

• Outstanding Paper Award, Association for Computational Linguistics (2023)

• Best Methods Paper Award, Nations of the Americas Chapter of the Association for Computational Linguistics (2022)

• Postgraduate Scholarship PGSD, NSERC (2018)