Chirag Nagpal

Understanding challenges to the interpretation of disaggregated evaluations of algorithmic fairness

Rewarding Progress: Scaling Automated Process Verifiers for LLM Reasoning

InfAlign: Inference-aware language model alignment

Theoretical guarantees on the best-of-n alignment policy

Helping or Herding? Reward Model Ensembles Mitigate but do not Eliminate Reward Hacking

Robust Preference Optimization through Reward Model Distillation