Salvato in:
| Autori principali: | , , , , , , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2026
|
| Soggetti: | |
| Accesso online: | https://arxiv.org/abs/2603.03988 |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866914368511279104 |
|---|---|
| author | Wang, Chunqi Wu, Bingchao Pang, Taotian Wang, Jiahao Yang, Jie Liu, Jia Zhang, Hao Zhu, Hai Shen, Lei Wang, Shizhun Wang, Bing Zeng, Xiaoyi |
| author_facet | Wang, Chunqi Wu, Bingchao Pang, Taotian Wang, Jiahao Yang, Jie Liu, Jia Zhang, Hao Zhu, Hai Shen, Lei Wang, Shizhun Wang, Bing Zeng, Xiaoyi |
| contents | 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%). |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_03988 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | SORT: A Systematically Optimized Ranking Transformer for Industrial-scale Recommenders Wang, Chunqi Wu, Bingchao Pang, Taotian Wang, Jiahao Yang, Jie Liu, Jia Zhang, Hao Zhu, Hai Shen, Lei Wang, Shizhun Wang, Bing Zeng, Xiaoyi Information Retrieval 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%). |
| title | SORT: A Systematically Optimized Ranking Transformer for Industrial-scale Recommenders |
| topic | Information Retrieval |
| url | https://arxiv.org/abs/2603.03988 |