Xiaoyi Liu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Xiaoyi Liu has a total of 46 co-authors. Here are the top co-authors by number of shared publications:
- Chunzhi Yang
Meta Platforms, Inc, Menlo Park, USA
2 shared publications - Huayu Li
Meta Platforms, Inc., Menlo Park, CA, USA
2 shared publications - Jade Nie
OpenAI, San Francisco, CA, USA
2 shared publications - Jiaqi Xu
Meta Platforms, Inc., Menlo Park, CA, USA
2 shared publications - Jiyan Yang
Meta Platforms, Inc., Menlo Park, CA, USA
2 shared publications - Laming Chen
Meta Platforms, Inc., Menlo Park, CA, USA
2 shared publications
Latest Publications
- Open Access
- 8 August 2026
- Published by Association for Computing Machinery (ACM)
- p. 7379-7390
Kunlun: Establishing Scaling Laws for Massive-Scale Recommendation Systems through Unified Architecture Design
Deriving predictable scaling laws that govern the relationship between model performance and computational investment is crucial for designing and allocating resources in massive-scale recommendation systems. While such laws are established for large language models, they remain challenging for recommendation systems, especially those processing both user history and context features. We identify poor scaling efficiency as the main barrier to predictable power-law scaling, stemming from inefficient modules with low Model FLOPs Utilization (MFU) and suboptimal resource allocation. We introduce Kunlun, a scalable architecture that systematically improves model efficiency and resource allocation. Our low-level optimizations include Generalized Dot-Product Attention (GDPA), Hierarchical Seed Pooling (HSP), and Sliding Window Attention. Our high-level innovations feature Computation Skip (CompSkip) and Event-level Personalization. These advances increase MFU from 17% to 37% on NVIDIA B200 GPUs and double scaling efficiency over state-of-the-art methods. Kunlun is now deployed in major Meta Ads models, delivering significant production impact.
- 8 August 2026
- Published by Association for Computing Machinery (ACM)
- p. 7379-7390
Open AccessKunlun: Establishing Scaling Laws for Massive-Scale Recommendation Systems through Unified Architecture Design
Deriving predictable scaling laws that govern the relationship between model performance and computational investment is crucial for designing and allocating resources in massive-scale recommendation systems. While such laws are established for large language models, they remain challenging for recommendation systems, especially those processing both user history and context features. We identify poor scaling efficiency as the main barrier to predictable power-law scaling, stemming from inefficient modules with low Model FLOPs Utilization (MFU) and suboptimal resource allocation. We introduce Kunlun, a scalable architecture that systematically improves model efficiency and resource allocation. Our low-level optimizations include Generalized Dot-Product Attention (GDPA), Hierarchical Seed Pooling (HSP), and Sliding Window Attention. Our high-level innovations feature Computation Skip (CompSkip) and Event-level Personalization. These advances increase MFU from 17% to 37% on NVIDIA B200 GPUs and double scaling efficiency over state-of-the-art methods. Kunlun is now deployed in major Meta Ads models, delivering significant production impact.
- Top Cited PapersOpen Access
- 10 November 2025
- Published by Association for Computing Machinery (ACM)
- p. 6225-6233
InterFormer: Effective Heterogeneous Interaction Learning for Click-Through Rate Prediction
- 10 November 2025
- Published by Association for Computing Machinery (ACM)
- p. 6225-6233
Top Cited PapersOpen AccessInterFormer: Effective Heterogeneous Interaction Learning for Click-Through Rate Prediction