SOTA Papers
LatestTrendingDigestSearch
Newsletter
  1. Home
  2. Longhao Jin

Longhao Jin

1 article on SOTA Papers

Computer Science

Request-Level Sharing Cuts Recommendation Inference Cost

ROCS separates request-side and candidate-side computation, reaching up to 3× higher retrieval QPS and 50% higher ranking QPS in production.

cs.AIcs.IRcs.LG
Aug 1, 20265 min2607.27744

Co-authors

Yuxin ChenLiang LuoBuyun ZhangJian JiaoBoda LiHaoyu WangTongyi TangAo CaiZijian ShenZhengkai ZhangWenyi XieRyan DickHan LiuNeng ShiBin YuJianbo XiaoShuyao BiHongtao YuYuanwei FangZhuoran ZhaoSijia ChenYang ChenShuqi YangQianru LiZikun LiuWei LingSihan ZengJiaxin LuYinbin MaJiawei LiYichen RuanYong Ler LeeBirmingham GuanZijian LiJianbo SunZhengyu ZhangZeliang ChenXiaohan WeiYuchen HaoGP MusumeciVenkatesh RanganathanYantao YaoChunqiang TangWenlin ChenSantanu KolayEllie Dingqiao Wen
SOTA Papers

Curated editorial synthesis of meaningful arXiv papers. No slop, no overclaiming.

Fields

  • Computer Science
  • Physics
  • Mathematics
  • Economics
  • Electrical Engineering

Navigation

  • Latest
  • Trending
  • Weekly Digest
  • Search
  • Contact

Legal

  • Terms of Service
  • Privacy Policy

© 2026 SOTA Papers. All rights reserved.

Data sourced from arXiv.org