解决商品推荐中稀疏词汇的表达失效问题，提升模型可扩展性。
OCP: Orthogonal Constrained Projection for Sparse Scaling in Industrial Commodity Recommendation
- 引入正交约束投影，优化嵌入表示的结构稳定性。
- 在京东大规模部署中实现UCXR提升12.97%、GMV增长8.9%。
- 适合需要扩展稀疏词汇和密集网络的工业级推荐系统。
在工业商品推荐系统中，物品ID词表的表示质量直接影响模型的可扩展性与泛化能力。传统物品ID词表在稀疏缩放时易受低频信息干扰，限制其对海量物品的表达能力，并导致表示坍塌。为此，本文提出正交约束投影（OCP）方法，通过强制正交性，约束反向传播流形，使学习到的嵌入奇异值谱与正交基对齐。该对齐保证高奇异熵，从而保留各向同性的泛化特征，抑制虚假相关性和对罕见物品的过拟合。实验表明，OCP加速损失收敛并增强模型可扩展性；尤其在扩展密集层时表现稳定。在京东的规模化部署验证了其有效性：UCXR提升12.97%，GMV增长8.9%，充分证明其在稀疏词表与密集架构扩展中的强实用性。
原文摘要 · Abstract (English)
In industrial commodity recommendation systems, the representation quality of Item-Id vocabularies directly impacts the scalability and generalization ability of recommendation models. A key challenge is that traditional Item-Id vocabularies, when subjected to sparse scaling, suffer from low-frequency information interference, which restricts their expressive power for massive item sets and leads to representation collapse. To address this issue, we propose an Orthogonal Constrained Projection method to optimize embedding representation. By enforcing orthogonality, the projection constrains the backpropagation manifold, aligning the singular value spectrum of the learned embeddings with the orthogonal basis. This alignment ensures high singular entropy, thereby preserving isotropic generalized features while suppressing spurious correlations and overfitting to rare items. Empirical results demonstrate that OCP accelerates loss convergence and enhances the model's scalability; notably, it enables consistent performance gains when scaling up dense layers. Large-scale industrial deployment on JD.com further confirms its efficacy, yielding a 12.97% increase in UCXR and an 8.9% uplift in GMV, highlighting its robust utility for scaling up both sparse vocabularies and dense architectures.
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。