Hong Yan
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Hong Yan has a total of 40 co-authors. Here are the top co-authors by number of shared publications:
- Hong Li
Meta, Menlo Park, CA, USA
6 shared publications - Junjie Yang
Meta, Menlo Park, USA
4 shared publications - Li Yu
Meta, Menlo Park, USA
3 shared publications - Renzhi Wu
Meta, Menlo Park, CA, USA
3 shared publications - Sri Reddy
Meta Platforms, Inc., Menlo Park, CA, USA
3 shared publications - Dongqi Fu
MRS, Meta, Menlo Park, CA, USA
2 shared publications
Latest Publications
- Open Access
- 27 September 2026
- Published by Association for Computing Machinery (ACM)
- p. 1515-1524
Structuring and Tokenizing Distributed User Interest Context for Generative Recommendation
- 27 September 2026
- Published by Association for Computing Machinery (ACM)
- p. 1515-1524
Open AccessStructuring and Tokenizing Distributed User Interest Context for Generative Recommendation
- Open Access
- 27 September 2026
- Published by Association for Computing Machinery (ACM)
- p. 1417-1426
RankGraph-2: Lifecycle Co-Design for Billion-Node Graph Learning in Recommendation
- 27 September 2026
- Published by Association for Computing Machinery (ACM)
- p. 1417-1426
Open AccessRankGraph-2: Lifecycle Co-Design for Billion-Node Graph Learning in Recommendation
- Open Access
- 19 July 2026
- Published by Association for Computing Machinery (ACM)
- p. 4545-4549
CMSL: Constructive Multi-Sequence Learning for Recommendation Systems
- 19 July 2026
- Published by Association for Computing Machinery (ACM)
- p. 4545-4549
Open AccessCMSL: Constructive Multi-Sequence Learning for Recommendation Systems
- Open Access
- 21 February 2026
- Published by Association for Computing Machinery (ACM)
- p. 1345-1346
RankGraph-Context: Empowering Different Industrial Recommendation System Stages
- 21 February 2026
- Published by Association for Computing Machinery (ACM)
- p. 1345-1346
Open AccessRankGraph-Context: Empowering Different Industrial Recommendation System Stages
- Open Access
- 22 September 2025
- Published by Association for Computing Machinery (ACM)
- p. 1050-1053
RankGraph: Unified Heterogeneous Graph Learning for Cross-Domain Recommendation
- 22 September 2025
- Published by Association for Computing Machinery (ACM)
- p. 1050-1053
Open AccessRankGraph: Unified Heterogeneous Graph Learning for Cross-Domain Recommendation
- Open Access
- 8 October 2024
- Published by Association for Computing Machinery (ACM)
- p. 778-780
Enhancing Performance and Scalability of Large-Scale Recommendation Systems with Jagged Flash Attention
The integration of hardware accelerators has significantly advanced the capabilities of modern recommendation systems, enabling the exploration of complex ranking paradigms previously deemed impractical. However, the GPU-based computational costs present substantial challenges. In this paper, we demonstrate our development of an efficiency-driven approach to explore these paradigms, moving beyond traditional reliance on native PyTorch modules. We address the specific challenges posed by ranking models’ dependence on categorical features, which vary in length and complicate GPU utilization. We introduce Jagged Feature Interaction Kernels, a novel method designed to extract fine-grained insights from long categorical features through efficient handling of dynamically sized tensors. We further enhance the performance of attention mechanisms by integrating Jagged tensors with Flash Attention. Our novel Jagged Flash Attention achieves up to 9 × speedup and 22 × memory reduction compared to dense attention. Notably, it also outperforms dense flash attention, with up to 3 × speedup and 53% more memory efficiency. In production models, we observe 10% QPS improvement and 18% memory savings, enabling us to scale our recommendation systems with longer features and more complex architectures.
- 8 October 2024
- Published by Association for Computing Machinery (ACM)
- p. 778-780
Open AccessEnhancing Performance and Scalability of Large-Scale Recommendation Systems with Jagged Flash Attention
The integration of hardware accelerators has significantly advanced the capabilities of modern recommendation systems, enabling the exploration of complex ranking paradigms previously deemed impractical. However, the GPU-based computational costs present substantial challenges. In this paper, we demonstrate our development of an efficiency-driven approach to explore these paradigms, moving beyond traditional reliance on native PyTorch modules. We address the specific challenges posed by ranking models’ dependence on categorical features, which vary in length and complicate GPU utilization. We introduce Jagged Feature Interaction Kernels, a novel method designed to extract fine-grained insights from long categorical features through efficient handling of dynamically sized tensors. We further enhance the performance of attention mechanisms by integrating Jagged tensors with Flash Attention. Our novel Jagged Flash Attention achieves up to 9 × speedup and 22 × memory reduction compared to dense attention. Notably, it also outperforms dense flash attention, with up to 3 × speedup and 53% more memory efficiency. In production models, we observe 10% QPS improvement and 18% memory savings, enabling us to scale our recommendation systems with longer features and more complex architectures.