发表机构
LinkedIn（领英）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
提出嵌入子空间划分（ESP）检索框架，通过分解嵌入为任务感知子空间并加权求和，实现服务时可调的多目标动态检索，在MS MARCO基准和LinkedIn平台上显著优于多任务基线。
AI 中文摘要
现代工业推荐系统必须在相互竞争的目标之间进行优化，在语义相关性与参与度和收入等业务指标之间取得平衡。虽然双编码器因其高效性主导了大规模检索，但它们将这些异构信号压缩到一个单一的静态嵌入空间中。这种设计造成了一个根本性的限制：一旦训练完成，检索器无法在不重新训练的情况下在服务时适应不断变化的目标优先级。此外，使用多目标损失进行联合优化通常会导致目标之间的干扰，从而产生次优的权衡。我们提出了嵌入子空间划分（ESP），这是一种检索框架，它将嵌入分解为任务感知的子空间，并用每个子空间相似度的加权和取代单一的点积，其权重可在服务时调整。对于Transformer双编码器，ESP使用模型原生的序列结束标记作为分段分隔符，通过分段感知的注意力掩码和位置编码重置，在单次前向传播中保证子空间隔离。服务通过在一个拼接索引上进行GPU加速的穷举kNN执行，消除了多头方法所需的每个目标的近似最近邻（ANN）基础设施。我们在基于MS MARCO构建的开源基准上评估了ESP。一个单一的ESP模型描绘了广泛的帕累托前沿，在不同的操作点上始终优于强大的多任务基线。在LinkedIn的职位匹配平台（每周超过7000万用户）上，ESP实现了动态检索重新配置，并带来了显著的关键业务指标提升。
英文摘要
Modern industrial recommender systems must optimize across competing objectives, balancing semantic relevance with business metrics such as engagement and revenue. While bi-encoders dominate large-scale retrieval due to their efficiency, they collapse these heterogeneous signals into a single static embedding space. This design creates a fundamental limitation: once trained, the retriever cannot adapt to shifting objective priorities at serving time without retraining. Moreover, joint optimization with multi-objective losses often induces interference between objectives, leading to suboptimal trade-offs. We propose Embedding Subspace Partitioning (ESP), a retrieval framework that decomposes the embedding into task-aware subspaces and replaces the single dot product with a weighted sum of per-subspace similarities, whose weights are tunable at serving time. For Transformer bi-encoders, ESP uses the model's native end-of-sequence token as a segment delimiter, with segment-aware attention masking and position encoding resets to guarantee subspace isolation in a single forward pass. Serving is performed via GPU-accelerated exhaustive kNN over one concatenated index, eliminating the need for per-objective Approximate Nearest Neighbor (ANN) infrastructure required by multi-head approaches. We evaluate ESP on an open-source benchmark built from MS MARCO. A single ESP model traces a broad Pareto frontier, consistently outperforming strong multi-task baselines across diverse operating points. In LinkedIn's job matching platform (70M+ weekly users), ESP enabled dynamic retrieval reconfiguration and delivered significant key business metric lifts.
Comments10 pages. To appear in the 20th ACM Conference on Recommender Systems (RecSys 2026)
