5 papers
Co-Designing Graph-based Approximate Nearest Neighbor Search at Billion Scale for Processing-in-Memory
Sitian Chen, Yusen Li, Yao Chen +3
Approximate Nearest Neighbor Search (ANNS) is a core primitive in modern AI systems, and graph-based methods currently offer the best accuracy-efficiency trade-off at scale. The wo…
Near-Zero-Overhead Freshness for Recommendation Systems via Inference-Side Model Updates
Wenjun Yu, Sitian Chen, Cheng Chen +1
Deep Learning Recommendation Models (DLRMs) underpin personalized services but face a critical freshness-accuracy tradeoff due to massive parameter synchronization overheads. Produ…
UpANNS: Enhancing Billion-Scale ANNS Efficiency with Real-World PIM Architecture
Sitian Chen, Amelie Chi Zhou, Yucheng Shi +2
Approximate Nearest Neighbor Search (ANNS) is a critical component of modern AI systems, such as recommendation engines and retrieval-augmented large language models (RAG-LLMs). Ho…
NM-SpMM: Accelerating Matrix Multiplication Using N:M Sparsity with GPGPU
Cong Ma, Du Wu, Zhelang Deng +11
Deep learning demonstrates effectiveness across a wide range of tasks. However, the dense and over-parameterized nature of these models results in significant resource consumption…
UpDLRM: Accelerating Personalized Recommendation using Real-World PIM Architecture
Sitian Chen, Haobin Tan, Amelie Chi Zhou +2
Deep Learning Recommendation Models (DLRMs) have gained popularity in recommendation systems due to their effectiveness in handling large-scale recommendation tasks. The embedding…