RecHarness: A Bandit-Routed Agentic Harness for Self-Evolving Recommender Systems

• 类型：arxiv
• 标识：2607.29241
• 链接：https://arxiv.org/abs/2607.29241
• 主分类：evaluation
• 形态：method
• 被引：0
• 被引来源：Semantic Scholar
• S2被引：0
• 影响力被引：0
• TLDR：RecHarness separates the optimization process into two steps: a bandit router selects the next modification direction according to historical validation feedback, while the LLM generates a concrete optimization hypothesis and executable code edit within the selected direction.
• 副分类：agent
• 待LLM分类：否
• 标题中文：RecHarness：用于自演化推荐系统的 Bandit 路由 Agentic Harness
• TLDR中文：RecHarness 将优化过程分为两步：bandit 路由器根据历史验证反馈选择下一步修改方向，LLM 在选定方向内生成具体优化假设与可执行代码编辑
• 来源文件：
• /inbox/tom/_candidates/2026-08-04-agent-rag-longcontext-candidates.json
• [S2 enrich]