Self-improving Prompting
摘要
What happens when prompts learn from their own output? This chapter examines how iterative refinement, feedback loops, and evaluation cues can improve performance over time—both for models and users.