Research includes: Connecting attention and modern Hopfield networks, developing plug-in associative memory modules for retrieval and editing, studying in-context learning via internal algorithm execution, and characterizing the universality of minimalist Transformers and prompt-based algorithm emulation.
Research Experience
Studies modern foundation models through four pillars: Learning, Storing, Computing, and Universality. Pursues research through neuroscience, statistics, information, and computation.
Education
PhD candidate in Computer Science at Northwestern University, advised by Han Liu; B.S. in Physics from National Taiwan University, advised by Pisin Chen.
Background
Research interests: Theoretical foundations and principled methodologies for large foundation models (e.g., Large Language Models and Generative AI). Long-term goal is to leverage machine learning to tackle important scientific and societal challenges.
Miscellany
Engages in interdisciplinary research collaborations, including Particle Physics at Fermilab, Drug Design at Abbvie, Finance at Gamma Paradigm Capital, and NdLinear & NdLinear-LoRA at Ensemble AI. Offers 2 hours weekly for individual support office hours, welcoming students from underrepresented groups to schedule a chat.