SORT: A Systematically Optimized Ranking Transformer for Industrial-scale Recommenders

TL;DR

SORT is a novel Transformer-based ranking model tailored for industrial-scale recommender systems, effectively addressing feature sparsity and low label density, and demonstrating significant improvements in business metrics and system efficiency.

Contribution

The paper introduces SORT, a scalable and optimized Transformer architecture for industrial ranking, with novel techniques to handle sparsity, stabilize training, and improve hardware efficiency.

Findings

Outperforms strong baselines in scalability and accuracy.

Achieves significant business metric improvements in e-commerce.

Halves latency and doubles throughput in deployment.

Abstract

While Transformers have achieved remarkable success in LLMs through superior scalability, their application in industrial-scale ranking models remains nascent, hindered by the challenges of high feature sparsity and low label density. In this paper, we propose SORT (Systematically Optimized Ranking Transformer), a scalable model designed to bridge the gap between Transformers and industrial-scale ranking models. We address the high feature sparsity and low label density challenges through a series of optimizations, including request-centric sample organization, local attention, query pruning and generative pre-training. Furthermore, we introduce a suite of refinements to the tokenization, multi-head attention (MHA), and feed-forward network (FFN) modules, which collectively stabilize the training process and enlarge the model capacity. To maximize hardware efficiency, we optimize our training system to elevate the model FLOPs utilization (MFU) to 22%. Extensive experiments demonstrate that SORT outperforms strong baselines and exhibits excellent scalability across data size, model size and sequence length, while remaining flexible at integrating diverse features. Finally, online A/B testing in large-scale e-commerce scenarios confirms that SORT achieves significant gains in key business metrics, including orders (+6.35%), buyers (+5.97%) and GMV (+5.47%), while simultaneously halving latency (-44.67%) and doubling throughput (+121.33%).