Expert cost in MoE serving follows a two-regime max-affine law, and a makespan-aware dispatcher over this model tracks the best fixed policy and wins where regimes mix.
MoonEP: A perfectly balanced expert parallelism library via dynamic redundant experts.https://github.com/ MoonshotAI/MoonEP, 2026
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
baseline 1
citation-polarity summary
fields
cs.DC 1years
2026 1verdicts
ACCEPT 1roles
baseline 1polarities
baseline 1representative citing papers
citing papers explorer
-
TEMPO: Makespan-Aware Expert-Parallel Load Balancing Across Memory- and Compute-Bound Regimes
Expert cost in MoE serving follows a two-regime max-affine law, and a makespan-aware dispatcher over this model tracks the best fixed policy and wins where regimes mix.