Paper 'When To Solve, When To Verify' accepted to CoLM 2025; 'VideoPhy-2' won the best paper award at ICML 2025; 'MathVista' recognized as the most influential ICLR 2024 paper; 'Stormer' won the best paper award at ICLR 2024; 'VideoCon' won the best paper award at ICLR 2024; 'CleanCLIP' accepted as an Oral Presentation at ICCV 2023 and won the best paper award at the Reliable and Trustworthy Large Scale ML workshop in ICLR 2023; 'GeoMLAMA' and 'ENTIGEN' accepted as Oral Presentations at EMNLP 2022; 'CyCLIP' accepted as an Oral Presentation at NeurIPS 2022.
Research Experience
Worked as a Student Researcher at Google DeepMind in Summer 2024 (mentors: Mehran Kazemi, Rishabh Agarwal, Vinh Q. Tran); interned at AWS AI in Summer 2022 (mentors: Karthik Gopalakrishnan, Sravan Bodapati); worked as an Analyst at Goldman Sachs for a year in 2021.
Education
Pursuing a Ph.D. at UCLA since 2021, co-advised by Kai-Wei Chang and Aditya Grover; completed B.Tech from IIIT Delhi in 2020, advised by Sumeet Agarwal.
Background
A Ph.D. candidate in the Department of Computer Science at UCLA, with research interests in reasoning in large language and multimodal models, vision-language understanding, and evaluating foundation models.
Miscellany
Actively collaborates with other UCLA faculty, including Nanyun Peng, Jeffrey Brantingham, Hongjing Lu; works with Google on multimodal learning (mentors: Yonatan Bitton, Idan Szpektor) and datacomp on data curation (advisor: Ludwig Schmidt).