Abhishek Nadgeri
CS PhD student at Stony Brook University, advised by Prof. Niranjan Balasubramanian.
Email: abhisheksanjee.nadgeri [at] stonybrook [dot] edu
Looking for Summer 2027 research internships in post-training and alignment.
Research
I want to build language models that learn what people actually want and can personalise according to them.
Interests: LLM post-training and alignment, reward modeling, robust evaluation, and AI safety.
Current work
Quality-aware RLVR for instruction following
When we train models to follow instructions, how do we keep the answer useful?