一个模型统一电商多种推荐场景，提升转化效果
OneMall: One Architecture, More Scenarios -- End-to-End Generative Recommender Family at Kuaishou E-Commerce
- 用统一架构整合商品卡、短视频、直播推荐
- 跨场景提升GMV最高+13.01%，订单量最高+15.32%
- 适合做多场景推荐系统落地的工程师和研究者
在生成式推荐浪潮中，我们提出OneMall，一个面向快手电商的端到端生成式推荐框架。该框架系统性地统一了商品卡、短视频、直播等多种商品分发场景。其核心包含三个部分：（1）电商语义分词器，捕捉真实语义与业务相关性；（2）基于Transformer的架构，采用Query-Former压缩长序列、交叉注意力融合多行为序列、稀疏MoE实现可扩展的自回归生成；（3）强化学习管道，通过奖励信号连接检索与排序模型，实现端到端策略优化。大量实验表明，OneMall在各场景均取得稳定提升：商品卡场景GMV提升13.01%，短视频场景订单量提升15.32%，直播场景订单量提升2.78%。目前已在快手上线，服务超4亿日活用户。
原文摘要 · Abstract (English)
In the wave of generative recommendation, we present OneMall, an end-to-end generative recommendation framework tailored for e-commerce services at Kuaishou. Our OneMall systematically unifies the e-commerce's multiple item distribution scenarios, such as Product-card, short-video and live-streaming. Specifically, it comprises three key components, aligning the entire model training pipeline to the LLM's pre-training/post-training: (1) E-commerce Semantic Tokenizer: we provide a tokenizer solution that captures both real-world semantics and business-specific item relations across different scenarios; (2) Transformer-based Architecture: we largely utilize Transformer as our model backbone, e.g., employing Query-Former for long sequence compression, Cross-Attention for multi-behavior sequence fusion, and Sparse MoE for scalable auto-regressive generation; (3) Reinforcement Learning Pipeline: we further connect retrieval and ranking models via RL, enabling the ranking model to serve as a reward signal for end-to-end policy retrieval model optimization. Extensive experiments demonstrate that OneMall achieves consistent improvements across all e-commerce scenarios: +13.01\% GMV in product-card, +15.32\% Orders in Short-Video, and +2.78\% Orders in Live-Streaming. OneMall has been deployed, serving over 400 million daily active users at Kuaishou.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。