MetaStrategy: Generative Ranking with Executable LLM Strategies
2026-08-10 • Information Retrieval
Information Retrieval
AI summaryⓘ
The authors propose MetaStrategy, a new approach for recommendation systems that creates detailed ranking plans instead of just ranking items directly. Their method uses a large language model to generate a JSON-based strategy controlling item selection and ranking rules, which can work alongside existing systems safely. They trained MetaStrategy by replaying past user requests without affecting real users, improving how recommendations are chosen. Tested on Taobao, their method improved user clicks, item views, and sales without slowing down the system.
recommender systemsranking strategylarge language modelJSONre-rankingonline A/B testpolicy trainingevaluatorproduction replayTaobao
Authors
Chengyu Lai, Jiuning Lin, Zhibo Xiao, Xiaodong Zhu, Ruiquan Lan, Bin Zhang, Zihong Huang, Wendong Zhang, Chuxin Chen, Yinjiang Cai, Shuai Zhong, Lingqing Zhang, Dimin Wang, Jialin Zhu, Han Zhu
Abstract
Industrial recommender systems rank heterogeneous content under coupled user, business, commercial, and experience objectives. Existing generative ranking methods typically construct item sequences directly, making them difficult to integrate with mature predictive models, operational rules, and field-level guardrails. We present MetaStrategy, a framework that instead generates a structured, executable ranking strategy. Conditioned on request context, a large language model (LLM) policy emits a typed JSON bundle controlling objective weights, content and category preferences, experience constraints, and position policies. A deterministic validator and compiler instantiate an isolated Generator that competes atomically with incumbents under the list-level Evaluator of the Generator-Evaluator (GE) architecture. We train the policy in a production-path replay environment that re-executes logged requests through the current re-ranking stack without user exposure. The method combines selection, relative-rank, and baseline-lift rewards, a self-competitive curriculum that feeds frequent strategies back as competitors, and Evaluator-routed reward-augmented on-policy distillation that transfers complementary 4B-parameter Teachers into a compact 0.8B-parameter Student. We deploy MetaStrategy in Taobao Homepage Guess You Like through diff-triggered nearline generation; LLM inference remains outside synchronous ranking, with no observable increase in response time (RT). In a seven-day user-randomized online A/B test, MetaStrategy wins 27.93% of treatment-side GE calls and significantly improves click page views (click PV) by 2.11%, item-detail page views (IPV) by 3.12%, and transaction amount by 2.83%.