Aligning LLM Agents by Learning Latent Preference from User Edits

Gao, Ge, Taymanov, Alexey, Salinas, Eduardo · arXiv · 2024