Ved Chadderwala
Graduate Student · Thomas Lord Department of Computer Science · University of Southern California
I'm a first-year MSCS student at USC focused on NLP, more precisely, inference-time evaluation of LLMs with a focus towards post-training with RL
Research Focus
Trustworthy LLMs · RL Post-training · Datasets
I study trustworthy machine learning and NLP, with an emphasis on reliability, robustness, and bias in real-world systems. My work spans inference-time reliability for LLMs, security critical ML for IoT, and ethical limitations of language-based user modeling.