用统一大模型实现推荐系统中用户与物品的无缝对齐
EGR: Embedding-Native Generative Retrieval with a Shared LLM
- 共享一个大模型同时学习物品和用户嵌入表示
- 在真实业务中提升2.91%转化率，冷启动效果佳
- 无需额外训练编码器，系统更简洁易维护
生成式召回在大规模推荐与广告系统中日益流行，但现有方法存在实际复杂性：语义ID方法依赖量化、可变的标识符词表及词元到物品的映射；基于嵌入的方法将物品编码器与查询生成器分开训练，限制了用户-物品对齐。我们提出EGR（Embedding-Native Generative Retrieval），一种面向推荐与广告的嵌入原生生成式召回框架。EGR使用单一共享的大语言模型，在同一嵌入空间中从物品元数据学习物品表示，从用户交互历史学习用户表示。物品直接以密集向量形式索引，用户历史被编码为密集检索查询。通过联合对比学习，将相关物品聚类，并使查询与其目标物品对齐。我们在公开基准、工业数据及线上部署中评估EGR。EGR在Amazon Reviews上优于已发表基线；在Snap DPA上，其性能随数据增长，能有效处理冷启动物品，并支持多模态输入。在生产环境中，EGR带来+2.91%的转化率提升，简化系统设计的同时提升了召回质量与广告表现。
原文摘要 · Abstract (English)
Generative retrieval is increasingly popular in large-scale recommendation and advertising systems, yet current methods introduce practical complications. Semantic-ID methods rely on quantization, mutable identifier vocabularies, and token-to-item grounding; embedding-based pipelines train the item encoder separately from the query generator, which limits user-item alignment. We propose EGR, an Embedding-Native Generative Retrieval framework for recommendation and advertising. EGR uses a single shared LLM to learn item representations from item metadata and user representations from interaction histories in one embedding space. Items are indexed directly as dense vectors, and user histories are encoded as dense retrieval queries. Joint contrastive training groups related items and aligns queries with their target items. We evaluate EGR on public benchmarks, industrial data, and live deployment. EGR outperforms published baselines on Amazon Reviews; on Snap DPA, it scales with data, handles cold-start items, and benefits from multimodal input. In production, EGR delivers a +2.91% conversion-rate lift, simplifying system design while improving retrieval quality and ad performance.
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。