作者位次 #25/29SIGIR 2026Oral· Information Retrieval (Full)
SilverTorch: A Unified Model-based System to Democratize Large-Scale Recommendation on GPUs
Bi Xue · Hong Wu · Lei Chen · Chao Yang · Yiming Ma · Fei Ding · Zhen Wang · Liang Wang · Xiaoheng Mao · Ke Huang · Xialu Li · Peng Xia · Rui Jian · Yanli Zhao · Yanzun Huang · Yijie Deng · Harry Tran · Ryan Chang · Min Yu · Eric Dong · Jiazhou Wang · Qianqian Zhang · Keke Zhai · Hongzhang Yin · Pawel Garbacki · Zheng Fang · Yiyi Pan · Min Ni · Yang Liu
摘要
Serving deep learning based recommendation models (DLRM) at scale is challenging. Existing approaches rely on dedicated ANN indexing and filtering services on CPUs, suffering from non-negligible costs and missing co-design opportunities. Such inefficiency makes them difficult to support complex m...