Multiple papers accepted at top-tier conferences: NeurIPS (2022, 2023 Spotlights), ICML (2023), CVPR XAI4CV Workshop (2023), FAccT (2024), WACV (2024 Oral), ICLR (2024 main conference and two workshops)
Developed WorldBench benchmark to uncover geographic disparities in LLMs
Proposed PRIME for interpretable failure model extraction
Introduced Text2Concept framework to derive concept activation vectors directly from text
Built practical tools for artistic copyright protection and improved T2I compositionality
Co-authored 'VLM Cookbook' survey (arXiv, 2024) with Meta FAIR researchers
Awarded Wylie Dissertation Fellowship (2024)
Received ARCS Foundation Fellowship from Metropolitan Washington Chapter (2023)
Named Outstanding Graduate Assistant at UMD (top 2%, 2023)
Invited talks at Google’s Theory and Practice of Foundation Models workshop (2024), Brown University’s Serre Lab (2023), and CMU AI Seminar Series (2023)