DREAM 技術報告

摘要

工業推薦系統通常採用級聯式的召回、排序與重排管線。儘管這些管線高效，但它們使資訊與目標在各模組間碎片化，依賴僵化規則，且對即時意圖的感知有限，導致瀏覽、比較與購買之間的會話級轉換未獲充分處理。我們提出DREAM（Developing Recommender Engine with Agentic Methods，即以智能體方法開發推薦引擎），這是一種自主最佳化控制架構，在不取代現有管線的前提下，於其上增加一層具感知能力、可編排且可審計的策略層。DREAM包含兩大核心組件。第一，三層意圖引擎（Intent Engine）將設備端信號融合為結構化的L0/L1/L2意圖表徵；其邊雲觸發鏈將上報量降至約8.7%。第二，元引擎（Meta Engine）利用元模型進行M1→M2→M3的分層推理：意圖摘要、由策略記憶（Strategy Memory）輔助的策略規劃，以及參數轉譯。元引擎透過具安全護欄的統一出入口分發所產生的參數。獎勵雙迴路（Reward Dual Loop）將離線模擬與線上回饋相結合——前者用於策略空間探索，後者用於結果校準——持續最佳化上述兩大組件，形成生成、執行、評估與經驗累積的閉環。淘寶首頁資訊流上的大規模A/B測試表明，僅對重排進行控制即可使IPV提升2.06%、核心IPV提升2.39%、GMV提升0.88%；將控制範圍擴展至精排後，上述增益分別提升至2.71%、3.06%與1.31%，同時PV持續提升超過1%。上述增益既無需更換管線模型，亦不損害服務穩定性，支持智能體式元控制作為工業推薦系統的一種可行範式。

English

Industrial recommender systems commonly use cascaded retrieval, ranking, and re-ranking pipelines. Although efficient, these pipelines fragment information and objectives across modules, rely on rigid rules, and have limited awareness of real-time intent, leaving session-level shifts among browsing, comparison, and purchase insufficiently addressed. We present DREAM (Developing Recommender Engine with Agentic Methods), an autonomous optimization control architecture that adds a perception-aware, orchestrable, and auditable policy layer atop existing pipelines without replacing them. DREAM has two core components. First, a three-tier Intent Engine fuses on-device signals into structured L0/L1/L2 intent representations; its edge-cloud trigger chain reduces reporting volume to approximately 8.7%. Second, a Meta Engine uses a MetaModel for layered M1-to-M2-to-M3 reasoning: intent summarization, strategy planning informed by Strategy Memory, and parameter translation. It dispatches the resulting parameters through a unified outlet with safety guardrails. A Reward Dual Loop continuously optimizes both components by combining offline simulation for strategy-space exploration with online feedback for outcome calibration, forming a cycle of generation, execution, evaluation, and experience accumulation. Large-scale A/B tests on Taobao's homepage feed show that re-ranking control alone improves IPV by 2.06%, Core IPV by 2.39%, and GMV by 0.88%. Extending control to fine ranking raises these gains to 2.71%, 3.06%, and 1.31%, respectively, while consistently improving PV by more than 1%. These gains require neither replacement of pipeline models nor compromise of serving stability, supporting agentic meta-control as a viable paradigm for industrial recommendation.