LoopCTR: Unlocking the Loop Scaling Power for Click-Through Rate Prediction
Jiakai Tang, Runfeng Zhang, Weiqiu Wang, Yifei Liu, Chuan Wang + 5 more
TLDR
LoopCTR uses a loop scaling paradigm to efficiently train high-performing CTR models, achieving SOTA with zero-loop inference.
Key contributions
- Introduces LoopCTR, a loop scaling paradigm that reuses shared layers to decouple computation from parameter growth.
- Employs a sandwich architecture with Hyper-Connected Residuals and Mixture-of-Experts.
- Uses process supervision at each loop depth, enabling a train-multi-loop, infer-zero-loop strategy.
- Achieves state-of-the-art CTR prediction performance on public and industrial datasets.
Why it matters
Traditional Transformer scaling for CTR models faces high overhead in industrial settings. LoopCTR addresses this by efficiently training powerful models without increasing inference complexity. This innovation makes advanced CTR prediction more practical and deployable.
Original Abstract
Scaling Transformer-based click-through rate (CTR) models by stacking more parameters brings growing computational and storage overhead, creating a widening gap between scaling ambitions and the stringent industrial deployment constraints. We propose LoopCTR, which introduces a loop scaling paradigm that increases training-time computation through recursive reuse of shared model layers, decoupling computation from parameter growth. LoopCTR adopts a sandwich architecture enhanced with Hyper-Connected Residuals and Mixture-of-Experts, and employs process supervision at every loop depth to encode multi-loop benefits into the shared parameters. This enables a train-multi-loop, infer-zero-loop strategy where a single forward pass without any loop already outperforms all baselines. Experiments on three public benchmarks and one industrial dataset demonstrate state-of-the-art performance. Oracle analysis further reveals 0.02--0.04 AUC of untapped headroom, with models trained with fewer loops exhibiting higher oracle ceilings, pointing to a promising frontier for adaptive inference.
📬 Weekly AI Paper Digest
Get the top 10 AI/ML arXiv papers from the week — summarized, scored, and delivered to your inbox every Monday.