Kushal Arora

language model (3) reinforcement learning (2) classification head (2) text generation (2) conditional generation (2) language modeling (1) supervised learning (1) imitation learning (1) toxicity detection (1) image captioning (1) self-supervised learning (1) multimodal learning (1) instruction following (1) sequence classification (1) model alignment (1) preference optimization (1) reward-based learning (1) reward learning (1) human feedback (1) direct preference optimization (1)