Yukti Makhija

reinforcement learning (1) preference optimization (1) language model alignment (1) reinforcement learning from human feedback (1) multiple instance learning (1) learning from label proportion (1) fine-grained scoring (1) span-level feedback (1) sentence-level scoring (1) reinforcement learning from ai feedback (1) sentence-level label (1) maximum likelihood calibration (1) label proportions learning (1)