RecHarness：用於自進化推薦系統的Bandit路由智能體框架

摘要

優化現代推薦模型仍然高度依賴工程師手動迭代架構、目標函數與訓練策略的變更。雖然基於大型語言模型（LLM）的智能體可以自動化這種試錯過程，但若讓LLM同時負責選擇修改方向並生成具體假設，往往會在有限的實驗預算下導致搜尋不穩定。受上述挑戰啟發，我們提出RecHarness——一種用於自動化推薦模型優化的Bandit路由智能體框架（Bandit-Routed Agentic Harness）。RecHarness將優化流程拆解為兩個步驟：Bandit路由器根據歷史驗證回饋選擇下一步的修改方向，而LLM則在該選定方向內生成具體的優化假設與可執行的程式碼修改。為了維持長程探索，RecHarness使用跳躍盆地機制，在局部修改停滯時啟動結構跳躍臂。在多個推薦任務、資料集與模型骨幹上，RecHarness比LLM推理搜尋取得更穩定的效能提升，並更有效地利用有限的實驗預算。在一場為期7天的大規模短視頻廣告平台線上A/B測試中，所選候選方案使ADVV提升2.084%、收入提升0.534%、曝光量提升0.559%。程式碼已公開於 https://github.com/6lyc/RecHarness。

English

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 propose RecHarness, a Bandit-Routed Agentic Harness for automated recommender model optimization. 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. 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 stable performance improvements and uses limited trial budgets more effectively than LLM-reasoning 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 https://github.com/6lyc/RecHarness.