Saved in:
Bibliographic Details
Main Authors: Sun, Chen, Xu, Beilin, Tan, Boheng, Wang, Jiacheng, Sun, Yuefeng, Bo, Rite, He, Ying, Zang, Yaqiang, Gong, Pinghua
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2603.18697
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914408888795136
author Sun, Chen
Xu, Beilin
Tan, Boheng
Wang, Jiacheng
Sun, Yuefeng
Bo, Rite
He, Ying
Zang, Yaqiang
Gong, Pinghua
author_facet Sun, Chen
Xu, Beilin
Tan, Boheng
Wang, Jiacheng
Sun, Yuefeng
Bo, Rite
He, Ying
Zang, Yaqiang
Gong, Pinghua
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2603_18697
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle OCP: Orthogonal Constrained Projection for Sparse Scaling in Industrial Commodity Recommendation
Sun, Chen
Xu, Beilin
Tan, Boheng
Wang, Jiacheng
Sun, Yuefeng
Bo, Rite
He, Ying
Zang, Yaqiang
Gong, Pinghua
Machine Learning
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.
title OCP: Orthogonal Constrained Projection for Sparse Scaling in Industrial Commodity Recommendation
topic Machine Learning
url https://arxiv.org/abs/2603.18697