Juntao Dai

large language model (4) safe reinforcement learning (3) constraint satisfaction (3) policy optimization (3) multimodal large language model (2) human preference alignment (2) preference learning (1) theorem proving (1) transfer learning (1) preference optimization (1) policy learning (1) visual reasoning (1) dialogue safety (1) ai safety (1) chain-of-thought reasoning (1) mathematical reasoning (1) formal verification (1) reinforcement learning from human feedback (1) error detection (1) question answering (1)