Huimin Xu

multi-task learning (2) process reward model (2) mathematical reasoning (2) reinforcement learning (2) direct preference optimization (1) self-supervised learning (1) attention mechanism (1) preference alignment (1) information extraction (1) named entity recognition (1) language model alignment (1) semantic consistency (1) reinforcement learning from human feedback (1) proximal policy optimization (1) human preference (1) syntactic dependency (1) reasoning chain (1) dependency tree (1) supervised machine learning (1) attribute value extraction (1)