Amrith Setlur

I’m a final year PhD Student in the Machine Learning Department at Carnegie Mellon University, where I am fortunate to be advised by Virginia Smith. I am also a long term visiting researcher at UC Berkeley, advised by Sergey Levine, and collaborating closely with Aviral Kumar. My PhD is generously supported by the JP Morgan AI PhD Fellowship award.

Research Overview

My research develops foundation models that adapt at deployment time by allocating test-time compute to reason, search, and interact with their environment, mitigating failures that arise when required behaviors are absent from training data. I study which training signals teach adaptation (including synthetic and model-generated trajectories) and which scalable algorithms best exploit verification and dense process rewards, drawing on meta-RL and reinforcement-learning-based post-training.