Lingjie Jiang

direct preference optimization (2) reinforcement learning (2) policy optimization (1) response generation (1) text generation (1) vision-language alignment (1) text-to-image generation (1) model alignment (1) diffusion model (1) multimodal large language model (1) task vector (1) human preference (1) language model fine-tuning (1) reward estimation (1) off-policy optimization (1) open-world environment (1) multilingual learning (1) large language model (1) natural language feedback (1) textual aesthetics (1)