提出TESR，让推荐系统更精准捕捉用户与候选项的细粒度互动。
Target-Aware Early Stage Ranking
- 引入注意力混合模块，建模用户历史与候选项的多层次匹配
- 在真实线上系统中实现离线指标提升和上线核心指标增长
- 支持大规模部署，首次在推荐早期阶段实现全目标感知序列建模
大规模推荐系统中的早期阶段排序（ESR）普遍采用双塔架构，虽可高效扩展，但难以直接捕捉用户与候选项之间的细粒度、目标感知交互。本文提出目标感知早期阶段排序（TESR），在双塔结构基础上引入注意力混合（MoA）模块，以请求级别序列建模用户历史。MoA包含：(i) 硬匹配注意力（HMA），捕获用户历史与候选项在类别ID层面的显式重叠信号；(ii) 针对候选项的隐式亲和性建模的靶向型HSTU注意力；(iii) 依赖与不依赖目标的交叉注意力，实现对称的用户-项目上下文化。在此基础上，通过多逻辑参数化门控（MLPG）头在打分时放大信号。为控制延迟在ESR预算内，联合设计了FP8量化、定制核函数及Torch Inductor编译路径。在实际生产部署中，TESR实现了稳定的离线NE提升与在线核心指标增长，据我们所知，是首个在如此规模下部署完整目标感知注意力序列建模的ESR系统。
原文摘要 · Abstract (English)
Early Stage Ranking (ESR) in large-scale recommendation systems is dominated by ''user--item decoupling'' Two Tower architectures, which scale efficiently but cannot capture fine-grained, target-aware user--item interactions directly. We propose Target-Aware Early Stage Ranking (TESR), which augments the Two Tower with a Mixture of Attention (MoA) module trained as a request-level sequence modeling over user history. MoA combines (i) Hard Matching Attention (HMA) to capture explicit categorical-ID level overlap signals between user history and candidate item, (ii) target-aware HSTU attention for implicit affinities conditioned on the candidate, and (iii) target dependent and independent cross-attention for symmetric user-item contextualization. On top of this, a Multi-Logit Parameterized Gating (MLPG) head amplifies these signals at scoring time. To keep latency within ESR budgets, we co-design the architecture with FP8 quantization, custom kernels, and a Torch Inductor compilation path. On a production deployment, TESR delivers consistent offline NE wins and online topline gains, and is, to our knowledge, the first deployment of full target-aware attention sequence modeling in an ESR stage at this scale.
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。