OATML

Kunal is an MSc by Research student in the Oxford Applied and Theoretical Machine Learning Group (OATML) under the supervision of Yarin Gal and Sebastian Farquhar. He is interested in the safe and robust deployment of foundation models in real-world scenarios, rooted in the understanding of human cognition and value systems.

Previously, Kunal worked on task ambiguity in humans and language models with Noah Goodman in the Stanford Computation and Cognition Lab and on the emergence of grounded representations in reinforcement learning via sequence modeling with Ellie Pavlick in the Brown LUNAR Lab. He is also interested in the application of language models in educational contexts, which he explored with Dora Demszky in the Stanford NLP Group. Kunal completed his undergraduate education in computer science and linguistics at Brown University.