跳到正文
arXiv cs.AI· O. Ozan Koyluoglu·· 1 天前AI 评分32

HSAD:用失败经验训练顾问模型,为执行智能体生成结构化建议

Hindsight Structured Advice Distillation: Steering Executor Agents with Advisors Trained on Their Failures and Task Characteristics

AI 导读

研究者提出 hindsight structured advice distillation(HSAD),让一个小型顾问模型读取输入并生成结构化建议,追加到冻结执行智能体的提示词中。

来源:arXiv cs.AI · arxiv.org