OneMall: One Architecture, More Scenarios -- End-to-End Generative Recommender Family at Kuaishou E-Commerce
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2026
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866915768313053184 |
|---|---|
| author | Zhang, Kun Zhang, Jingming Cheng, Wei Cheng, Yansong Zhang, Jiaqi Lu, Hao Zhang, Xu Gan, Haixiang Cao, Jiangxia Wang, Tenglong Zhang, Ximing Xia, Boyang Cai, Kuo Wang, Shiyao Dou, Hongjian Yu, Jinkai Wen, Mingxing Luo, Qiang Liang, Dongxu Lei, Chenyi Wang, Jun Liu, Runan Liu, Zhaojie Tang, Ruiming Gao, Tingting Liu, Shaoguo Ding, Yuqing Kong, Hui Li, Han Zhou, Guorui Ou, Wenwu Gai, Kun |
| author_facet | Zhang, Kun Zhang, Jingming Cheng, Wei Cheng, Yansong Zhang, Jiaqi Lu, Hao Zhang, Xu Gan, Haixiang Cao, Jiangxia Wang, Tenglong Zhang, Ximing Xia, Boyang Cai, Kuo Wang, Shiyao Dou, Hongjian Yu, Jinkai Wen, Mingxing Luo, Qiang Liang, Dongxu Lei, Chenyi Wang, Jun Liu, Runan Liu, Zhaojie Tang, Ruiming Gao, Tingting Liu, Shaoguo Ding, Yuqing Kong, Hui Li, Han Zhou, Guorui Ou, Wenwu Gai, Kun |
| contents | In the wave of generative recommendation, we present OneMall, an end-to-end generative recommendation framework tailored for e-commerce services at Kuaishou. Our OneMall systematically unifies the e-commerce's multiple item distribution scenarios, such as Product-card, short-video and live-streaming. Specifically, it comprises three key components, aligning the entire model training pipeline to the LLM's pre-training/post-training: (1) E-commerce Semantic Tokenizer: we provide a tokenizer solution that captures both real-world semantics and business-specific item relations across different scenarios; (2) Transformer-based Architecture: we largely utilize Transformer as our model backbone, e.g., employing Query-Former for long sequence compression, Cross-Attention for multi-behavior sequence fusion, and Sparse MoE for scalable auto-regressive generation; (3) Reinforcement Learning Pipeline: we further connect retrieval and ranking models via RL, enabling the ranking model to serve as a reward signal for end-to-end policy retrieval model optimization. Extensive experiments demonstrate that OneMall achieves consistent improvements across all e-commerce scenarios: +13.01\% GMV in product-card, +15.32\% Orders in Short-Video, and +2.78\% Orders in Live-Streaming. OneMall has been deployed, serving over 400 million daily active users at Kuaishou. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_21770 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | OneMall: One Architecture, More Scenarios -- End-to-End Generative Recommender Family at Kuaishou E-Commerce Zhang, Kun Zhang, Jingming Cheng, Wei Cheng, Yansong Zhang, Jiaqi Lu, Hao Zhang, Xu Gan, Haixiang Cao, Jiangxia Wang, Tenglong Zhang, Ximing Xia, Boyang Cai, Kuo Wang, Shiyao Dou, Hongjian Yu, Jinkai Wen, Mingxing Luo, Qiang Liang, Dongxu Lei, Chenyi Wang, Jun Liu, Runan Liu, Zhaojie Tang, Ruiming Gao, Tingting Liu, Shaoguo Ding, Yuqing Kong, Hui Li, Han Zhou, Guorui Ou, Wenwu Gai, Kun Information Retrieval In the wave of generative recommendation, we present OneMall, an end-to-end generative recommendation framework tailored for e-commerce services at Kuaishou. Our OneMall systematically unifies the e-commerce's multiple item distribution scenarios, such as Product-card, short-video and live-streaming. Specifically, it comprises three key components, aligning the entire model training pipeline to the LLM's pre-training/post-training: (1) E-commerce Semantic Tokenizer: we provide a tokenizer solution that captures both real-world semantics and business-specific item relations across different scenarios; (2) Transformer-based Architecture: we largely utilize Transformer as our model backbone, e.g., employing Query-Former for long sequence compression, Cross-Attention for multi-behavior sequence fusion, and Sparse MoE for scalable auto-regressive generation; (3) Reinforcement Learning Pipeline: we further connect retrieval and ranking models via RL, enabling the ranking model to serve as a reward signal for end-to-end policy retrieval model optimization. Extensive experiments demonstrate that OneMall achieves consistent improvements across all e-commerce scenarios: +13.01\% GMV in product-card, +15.32\% Orders in Short-Video, and +2.78\% Orders in Live-Streaming. OneMall has been deployed, serving over 400 million daily active users at Kuaishou. |
| title | OneMall: One Architecture, More Scenarios -- End-to-End Generative Recommender Family at Kuaishou E-Commerce |
| topic | Information Retrieval |
| url | https://arxiv.org/abs/2601.21770 |