Ramamohan Paturi

Reasoning in langauge models · token bias · chain of thought · verification of langauge model responses · Reinforcement learning · Mechanistic interpretability · Circuit complexity of deep learning