Showing cs.LGShow all
3 papers · 1 filter
cs.LG2026
ROCS: Request-Oriented Compute Sharing for Efficient Large-Scale Recommendation
Yuxin Chen, Liang Luo, Buyun Zhang +44
Modern recommendation models gain prediction quality by scaling feature-interaction and sequence modules, but production cost constraints cap how far systems can scale. In this wor…
cs.LG2026
LoKA: Low-precision Kernel Applications for Recommendation Models At Scale
Liang Luo, Yinbin Ma, Quanyu Zhu +21
Recent GPU generations deliver significantly higher FLOPs using lower-precision arithmetic, such as FP8. While successfully applied to large language models (LLMs), its adoption in…
cs.LG2026
Scaling Multi-Node Mixture-of-Experts Inference Using Expert Activation Patterns
Abhimanyu Bambhaniya, Geonhwa Jeong, Jason Park +6
Most recent state-of-the-art (SOTA) large language models (LLMs) use Mixture-of-Experts (MoE) architectures to scale model capacity without proportional per-token compute, enabling…