arXiv：cs.AI· Zhimeng Jiang, Yuening Li, Ruixiao Sun, Yuanzhen Lin, Kun Jin, Chuan He, Liang Liu·· 4 小时前
链维度扩展 CDS：面向工业短视频排序的通信高效循环式扩展
Chain Dimension Scaling: Communication-Efficient Loop-Style Scaling for Industrial Short-Video Ranking
AI 导读
研究者提出链维度扩展（CDS），一种与宽度和深度正交的扩展策略：重复应用参数共享的单一特征交互块，以迭代次数（链长）作为不增加参数的新扩展维度。在生产级大规模多任务短视频排序系统中，压缩模式下 CDS 离线质量与基线差距在 0.03% 以内，可训练参数减少 26.6%，训练吞吐提升 6.4%；质量扩展模式下质量随链长单调提升。
正文
Abstract:Industrial recommendation rankers keep scaling in width and depth to billions of parameters, but this increasingly incurs costs in parameter growth, distributed communication overhead, and training inefficiency. We introduce \underline{C}hain \underline{D}imension \underline{S}caling (CDS), a scaling strategy orthogonal to width and depth. CDS repeatedly applies a single feature-interaction block with parameters shared across iterations, so the number of iterations (the chain length) becomes a new scaling dimension that adds no parameters. A lightweight intra-chain residual connection stabilizes optimization and improves quality. CDS enables two regimes. In the \emph{compression regime}, shallow blocks with a longer chain match deeper baselines with fewer parameters and faster training; in the \emph{quality-scaling regime}, increasing chain length under a fixed parameter budget yields consistent offline and online gains, suggesting the chain dimension adds modeling capacity beyond parameter reuse. We evaluate CDS in a large-scale multi-task short-video ranking system in production. In the compression regime, CDS stays within $0.03\%$ of baseline offline quality while reducing trainable parameters by 26.6\% and improving training throughput by 6.4\%, with neutral-to-positive online participation. In the quality-scaling regime, quality improves monotonically with chain length. Wall-clock analysis shows the efficiency gain comes mainly not from reduced FLOPs but from lower cross-device synchronization overhead and reduced optimizer update cost. These results position parameter-shared iterative computation as a practical, communication-efficient scaling axis for industrial ranking systems.
| Subjects: | Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.16415 [cs.AI] |
| (or arXiv:2610.16415v1 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2610.16415 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Zhimeng Jiang [view email]
[v1]
Mon, 12 Oct 2026 08:49:29 UTC (815 KB)
来源：arXiv：cs.AI · arxiv.org