I am looking for a post-doc position fall of 2024. I am interested in mechanistic interpretability. If you see me as a good fit, please email me at ajyl[at]umich.edu.
Hello! I am a PhD candidate at the University of Michigan, working on machine learning + natural language processing. I am privileged to be a part of the LIT research group, being advised by Professor Rada Mihalcea.
Currently, I am interested in understanding the mechanisms of neural networks, and interpreting their inner representations.
I have been fortunate to collaborate with multiple teams in industry (Meta AI x2, Microsoft Research). My collaborators include Neel Nanda (DeepMind), Jason Weston, Sainbayar Sukhbaatar, Mike Lewis, Emily Dinan, Jing Xu (Meta AI), and Silviu Cucerzan (Microsoft Research).
Prior to my Ph.D. program, I led the Core AI R&D team at Clinc, where we built and deployed virtual assistants and conversational AI platforms for financial institutions.