Scholarlink

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.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Self-improving Prompting

  • 章节

  • 发表日期：

    2025年

查看全文
  • Jordi Vallverdú

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.