5 citations · 8 across the 18 of their papers we have counts for
Showing cs.DCShow all
3 papers · 1 filter
cs.DC2026
TEMPO: Makespan-Aware Expert-Parallel Load Balancing Across Memory- and Compute-Bound Regimes
Jie Li, Chenxin Jia, Jinliang Shen +5
In expert-parallel (EP) MoE serving, every layer synchronizes at the slowest GPU. Dispatchers balance token counts (EPLB, LPLB, UltraEP) or activated-expert counts (METRO), assumin…
cs.DC2026
X-Stage: Modeling Post-Issue Backpressure in GPU Communication--Computation Fusion
Jianwen Xian, Zhiyuan Xu, Yuchen Li +9
Fine-grained, device-initiated communication allows fused GPU kernels to issue remote stores directly from their compute pipelines, a pattern increasingly used in expert parallelis…
cs.DC2026
PROBE: Co-Balancing Computation and Communication in MoE Inference via Real-Time Predictive Prefetching
Qianchao Zhu, Xucheng Ye, Yuliang Liu +2
Mixture-of-Experts models have become a dominant architecture for scaling Large Language Models by activating only a sparse subset of experts per token. However, latency-critical M…