Metraipolis › archive › dc_f9b697f00c76

ACE: Adaptive Calibration-Free Expert Skipping for MoE-based LLMs — arXiv:2609.05228

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.05228
License (as archived)
CC0-1.0
Archived
2026-10-16T22:21:17.005Z
SHA-256 of the archived text
a1603fb15d7c86b8e847eaaa70aa72aecc0ee2939ae783f09c0bafec45faf439
Stable URL of this page
https://metraipolis.com/documento/dc_f9b697f00c76

How to cite

ACE: Adaptive Calibration-Free Expert Skipping for MoE-based LLMs — arXiv:2609.05228 — arXiv (en), https://arxiv.org/abs/2609.05228 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_f9b697f00c76 (Metraipolis archive dc_f9b697f00c76, sha256 a1603fb15d7c86b8e847eaaa70aa72aecc0ee2939ae783f09c0bafec45faf439)
BibTeX
@misc{dcf9b697f00c76,
  title = {ACE: Adaptive Calibration-Free Expert Skipping for MoE-based LLMs — arXiv:2609.05228},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.05228},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_f9b697f00c76; sha256 a1603fb15d7c86b8e847eaaa70aa72aecc0ee2939ae783f09c0bafec45faf439},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.05228. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

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 complementary components: 1) Global Spectral Proxy (GSP), which estimates global transformation capacity from the coupled gate, up, and down projections together with RMSNorm scaling; and 2) Router-Conditioned Refinement (RCR), which constructs expert-specific direction prototypes from centered router weights and evaluates expert responses along routing-preferred directions. During inference, ACE combines both estimates with runtime router gates and skips an expert slot only when both views identify it as low-contribution, while always retaining the top-1 expert. All expert statistics are computed offline, leaving only table lookups and lightweight scalar operations online. Extensive experiments across three MoE-based LLMs and eight benchmarks demonstrate that ACE consistently outperforms existing static and dynamic baselines, with increasingly pronounced advantages under aggressive expert skipping. For instance, at a 50% skipping ratio on Qwen3.6-35B-A3B, ACE reduces WikiText-2 perplexity by 7.96% and improves average downstream accuracy by 4.15 percentage points over the strongest competing method.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-07 in cs.AI).
Categorie arXiv: Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2609.05228 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-16T22:18:11.443Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-07?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).