跳到正文
arXiv cs.AI· Haoran Ling, Yuecheng Li, Zeyu Song, Jing Yao, Shuwen Kang, Songcheng Ren, Chi Lu, Wenjin Wu, Peng Jiang·· 4 小时前AI 评分34

RecHarness:用 Bandit 路由的智能体框架实现推荐系统自我进化

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

AI 导读

RecHarness 将推荐模型优化拆成两步:Bandit 路由器依据历史验证反馈选择修改方向,LLM 在该方向内生成具体优化假设与可执行代码改动,并用 jump-basin 机制在局部编辑停滞时激活结构跳变臂。

正文 · 原文

View PDF HTML (experimental)

Abstract:Optimizing recommender models remains heavily reliant on manual iteration over architectures, objectives, and training strategies. While LLM-based agents can automate this trial-and-error process, relying on LLMs to both select modification directions and formulate concrete hypotheses can lead to unstable search under limited trial budgets. Inspired by the above challenge, we propose RecHarness, a Bandit-Routed Agentic Harness for automated recommender model optimization. RecHarness decouples 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. To sustain long-horizon exploration, RecHarness uses a jump-basin mechanism to activate a structural-jump arm when local edits stagnate. Across multiple recommendation tasks, datasets, and model backbones, RecHarness achieves more consistent performance improvements and uses limited trial budgets more effectively than LLM-directed search. During a 7-day online A/B test on a large-scale short-video advertising platform, the selected candidate improves ADVV by 2.084%, Revenue by 0.534%, and Exposure by 0.559%. Code is available at: this https URL.
Comments: 17 pages, 3 figures
Subjects: Information Retrieval (cs.IR); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
Cite as: arXiv:2607.29241 [cs.IR]
  (or arXiv:2607.29241v2 [cs.IR] for this version)
  https://doi.org/10.48550/arXiv.2607.29241

arXiv-issued DOI via DataCite

Submission history

From: Yuecheng Li [view email]
[v1] Fri, 31 Jul 2026 10:15:38 UTC (722 KB)
[v2] Tue, 17 Nov 2026 11:49:18 UTC (2,645 KB)

来源:arXiv cs.AI · arxiv.org