Jiangxia Cao
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Jiangxia Cao has a total of 118 co-authors. Here are the top co-authors by number of shared publications:
- Zhaojie Liu
Kuaishou Technology, Beijing, China
11 shared publications - Guorui Zhou
Kuaishou Technology, Beijing, China
7 shared publications - Kun Gai
Kuaishou Technology, Beijing, China
7 shared publications - Ruochen Yang
Kuaishou Technology, Beijing, China
6 shared publications - Shuang Yang
Kuaishou Technology, Beijing, China
6 shared publications - Tingwen Liu
Institute of Information Engineering, Chinese Academy of Sciences, Beijing, China and School of Cyber Security, University of Chinese Academy of Sciences, Beijing, China
5 shared publications
Latest Publications
- Open Access
- 6 November 2026
- Published by Association for Computing Machinery (ACM)
- p. 8688-8695
SARM: LLM-Augmented Semantic Anchor for End-to-End Live-Streaming Ranking
- 6 November 2026
- Published by Association for Computing Machinery (ACM)
- p. 8688-8695
Open AccessSARM: LLM-Augmented Semantic Anchor for End-to-End Live-Streaming Ranking
- Open Access
- 6 November 2026
- Published by Association for Computing Machinery (ACM)
- p. 8768-8775
OneMall: One Architecture, More Scenarios — End-to-End Generative Recommender Family for E-commerce
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: +14.7% GMV in Product-Card, +10.3% GMV in Short-Video, and +4.9% GMV in Live-Streaming. OneMall has been deployed, serving over 400 million daily active users at Kuaishou.
- 6 November 2026
- Published by Association for Computing Machinery (ACM)
- p. 8768-8775
Open AccessOneMall: One Architecture, More Scenarios — End-to-End Generative Recommender Family for E-commerce
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: +14.7% GMV in Product-Card, +10.3% GMV in Short-Video, and +4.9% GMV in Live-Streaming. OneMall has been deployed, serving over 400 million daily active users at Kuaishou.
- Open Access
- 6 November 2026
- Published by Association for Computing Machinery (ACM)
- p. 7568-7572
Distilling Post-Conversion Content for User Retention Modeling
- 6 November 2026
- Published by Association for Computing Machinery (ACM)
- p. 7568-7572
Open AccessDistilling Post-Conversion Content for User Retention Modeling
- Open Access
- 8 August 2026
- Published by Association for Computing Machinery (ACM)
- p. 8155-8166
OneLive: Dynamically Unified Generative Framework for Live-Streaming Recommendation
- 8 August 2026
- Published by Association for Computing Machinery (ACM)
- p. 8155-8166
Open AccessOneLive: Dynamically Unified Generative Framework for Live-Streaming Recommendation
- Open Access
- 20 April 2026
- Published by Association for Computing Machinery (ACM)
- p. 1775-1786
From Agnostic to Specific: Latent Preference Diffusion for Multi-Behavior Sequential Recommendation
- 20 April 2026
- Published by Association for Computing Machinery (ACM)
- p. 1775-1786
Open AccessFrom Agnostic to Specific: Latent Preference Diffusion for Multi-Behavior Sequential Recommendation
- Open Access
- 20 April 2026
- Published by Association for Computing Machinery (ACM)
- p. 2302-2311
FARM: Frequency-Aware Model for Cross-Domain Live-Streaming Recommendation
- 20 April 2026
- Published by Association for Computing Machinery (ACM)
- p. 2302-2311
Open AccessFARM: Frequency-Aware Model for Cross-Domain Live-Streaming Recommendation
- Open Access
- 21 February 2026
- Published by Association for Computing Machinery (ACM)
- p. 1078-1082
Foresight Prediction Enhanced Live-Streaming Recommendation
- 21 February 2026
- Published by Association for Computing Machinery (ACM)
- p. 1078-1082
Open AccessForesight Prediction Enhanced Live-Streaming Recommendation
- Top Cited PapersOpen Access
- 21 February 2026
- Published by Association for Computing Machinery (ACM)
- p. 1115-1119
ChorusCVR: Chorus Supervision for Entire Space Post-Click Conversion Rate Modeling
- 21 February 2026
- Published by Association for Computing Machinery (ACM)
- p. 1115-1119
Top Cited PapersOpen AccessChorusCVR: Chorus Supervision for Entire Space Post-Click Conversion Rate Modeling
- Open Access
- 21 February 2026
- Published by Association for Computing Machinery (ACM)
- p. 1073-1077
PushGen: Push Notifications Generation with LLM
- 21 February 2026
- Published by Association for Computing Machinery (ACM)
- p. 1073-1077
Open AccessPushGen: Push Notifications Generation with LLM
- Top Cited PapersOpen Access
- 10 November 2025
- Published by Association for Computing Machinery (ACM)
- p. 5915-5922
QARM: Quantitative Alignment Multi-Modal Recommendation at Kuaishou
- 10 November 2025
- Published by Association for Computing Machinery (ACM)
- p. 5915-5922
Top Cited PapersOpen AccessQARM: Quantitative Alignment Multi-Modal Recommendation at Kuaishou