Climber: Toward Efficient Scaling Laws for Large Recommendation Models

TL;DR

Climber introduces a scalable, efficient recommendation framework with novel architecture and acceleration techniques, enabling better model scaling and online performance without high resource costs, validated through extensive experiments and deployment.

Contribution

The paper presents a new recommendation model architecture with multi-scale sequence extraction and dynamic attention modulation, along with acceleration methods, to improve scaling and efficiency.

Findings

Achieves 5.15× throughput gain without performance loss.

Demonstrates a 12.19% online metric lift in real deployment.

Validates superior scaling behavior across multiple datasets.

Abstract

Transformer-based generative models have achieved remarkable success across domains with various scaling law manifestations. However, our extensive experiments reveal persistent challenges when applying Transformer to recommendation systems: (1) Transformer scaling is not ideal with increased computational resources, due to structural incompatibilities with recommendation-specific features such as multi-source data heterogeneity; (2) critical online inference latency constraints (tens of milliseconds) that intensify with longer user behavior sequences and growing computational demands. We propose Climber, an efficient recommendation framework comprising two synergistic components: the model architecture for efficient scaling and the co-designed acceleration techniques. Our proposed model adopts two core innovations: (1) multi-scale sequence extraction that achieves a time complexity reduction by a constant factor, enabling more efficient scaling with sequence length; (2) dynamic temperature modulation adapting attention distributions to the multi-scenario and multi-behavior patterns. Complemented by acceleration techniques, Climber achieves a 5.15$\times$ throughput gain without performance degradation by adopting a "single user, multiple item" batched processing and memory-efficient Key-Value caching. Comprehensive offline experiments on multiple datasets validate that Climber exhibits a more ideal scaling curve. To our knowledge, this is the first publicly documented framework where controlled model scaling drives continuous online metric growth (12.19\% overall lift) without prohibitive resource costs. Climber has been successfully deployed on Netease Cloud Music, one of China's largest music streaming platforms, serving tens of millions of users daily.