world knowledge tree

self-reflection refinement (1) llm alignment (1) iterative self-improvement (1) knowledge tree (1) world knowledge (1) supervised fine-tuning (1) synthetic data (1) reinforcement learning from human feedback (1) synthetic data generation (1)