SOURCE-LINKED INTELLIGENCE

ACE: Adaptive Calibration-Free Expert Skipping for MoE-based LLMs

arXiv · AI, language, vision and robotics · article · Sep 4, 2026 · UTC

Mixture-of-Experts (MoE) architectures provide an efficient paradigm for scaling large language models (LLMs), yet fixed top-k routing activates the same number of expert slots for every token, causing substantial redundant computation. Existing expert-skipping methods often rely on router confidence, calibration data, or additional training, and therefore cannot reliably estimate the actual contribution of routed experts. To this end, we propose ACE, a training-free, calibration-free, and checkpoint-preserving framework for token-adaptive expert skipping in MoE-based LLMs. ACE contains two co

Read original source ↗ Open in workspace

recordType

paper

region

Global