RecHarness：基于Bandit路由的自主进化推荐系统智能体框架

AI 摘要

RecHarness 提出了一种基于 Bandit 路由的智能体框架，用于自动化推荐模型优化。该框架将优化过程分为两步：Bandit 路由器根据历史验证反馈选择修改方向，LLM 在选定方向内生成具体优化假设和可执行代码编辑。在多个推荐任务和数据集上，RecHarness 比纯 LLM 推理搜索更稳定且更节省试验预算，在大型短视频广告平台的 7 天在线 A/B 测试中，ADVV 提升 2.084%，Revenue 提升 0.534%，Exposure 提升 0.559%。

正文节选

RecHarness: A Bandit-Routed Agentic Harness for Self-Evolving Recommender Systems Abstract Optimizing modern recommender models still depends heavily on engineers manually iterating over architectural, objective, and training-strategy changes. While LLM-based agents can automate this trial-and-error process, allowing the LLM to both select modification directions and generate concrete hypotheses often leads to unstable search under limited experiment budgets. Inspired by the above challenge, we

发布时间：—

抓取时间：2026-08-04 17:28

来源机构：Hugging Face