Parsa Mahmoudieh

I am currently working on intersections of Reinforcement Learning, NLP, and Computer Vision research at Google Deepmind in Mountain View. My team's latest work on LearnLM is preferred substantially by expert raters across a diverse set of learning scenarios, with average preference strengths of 31% over GPT-4o, 11% over Claude 3.5, and 13% over the Gemini 1.5 Pro model LearnLM was based on.

We integrated our latest work into the SFT, RM, and RL stages of Gemini 2.0. Our previous iteration of LearnLM was presented at Google I/O 2024 here! My main role has been in co-leading the RM & RL stages for LearnLM.

Previously I received my CS PhD at UC Berkeley in BAIR advised by Trevor Darrell and have been mentored by Deepak Pathak and Evan Shelhamer. I have also worked with Pulkit Agrawal, Alyosha Efros, and Jitendra Malik.

Before grad school, I did a double major in EECS and MechE at UC Berkeley and had done undergraduate research in Ron Fearing's Robotics lab. I've also had the pleasure to do internships at GM and Ford.

Email  /  Google Scholar  /  LinkedIn  /  PhD Thesis

Template from Jon Barron