Metraipolis › archive › dc_7cc52949308b
AirMoE: Realizing Over-the-Air Distributed Mixture-of-Experts Inference at the Wireless Edge — arXiv:2608.22932
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.22932
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-18T10:12:51.039Z
- SHA-256 of the archived text
8e6d345e4b8db10834c7a6cf6f77811c2fb444d871488cd69000ca4e07091d45- Stable URL of this page
https://metraipolis.com/documento/dc_7cc52949308b
How to cite
AirMoE: Realizing Over-the-Air Distributed Mixture-of-Experts Inference at the Wireless Edge — arXiv:2608.22932 — arXiv (en), https://arxiv.org/abs/2608.22932 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_7cc52949308b (Metraipolis archive dc_7cc52949308b, sha256 8e6d345e4b8db10834c7a6cf6f77811c2fb444d871488cd69000ca4e07091d45)
BibTeX
@misc{dc7cc52949308b,
title = {AirMoE: Realizing Over-the-Air Distributed Mixture-of-Experts Inference at the Wireless Edge — arXiv:2608.22932},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.22932},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_7cc52949308b; sha256 8e6d345e4b8db10834c7a6cf6f77811c2fb444d871488cd69000ca4e07091d45},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.22932. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Mixture-of-experts (MoE) architectures enable efficient large language model (LLM) inference at the wireless edge through sparse activation. The wireless distributed MoE (WIDE) architecture addresses edge-resource constraints by distributing experts across devices coordinated by an edge server. However, WIDE suffers from repeated uplink transmissions of high-dimensional expert outputs via orthogonal multiple access. To overcome this bottleneck, we propose AirMoE, an over-the-air computing (AirComp)-enabled framework for simultaneous expert-output aggregation via wireless waveform superposition. Integrating AirComp into MoE inference introduces three challenges: fast-varying aggregation weights, layer-dependent error sensitivity, and channel-aware expert placement. To address these challenges, we construct an inference-aware AirMoE error metric to quantify aggregation distortion effects on end-to-end (E2E) inference accuracy via perturbation-based layer-sensitivity calibration. We then formulate a joint optimization problem to minimize this error and decompose it, without loss of optimality, into a two-timescale framework. At the fast timescale, we derive a globally optimal threshold-based power-control policy that separates devices into coefficient-aligned and full-power groups. At the slow timescale, we develop an activation- and channel-aware expert placement strategy that assigns more important experts to devices with lower channel-power cost. Extensive experiments demonstrate that AirMoE outperforms representative baselines in E2E inference accuracy, especially under strong device heterogeneity. Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-08-25 in eess.SP). Categorie arXiv: Signal Processing (eess.SP). Pagina del lavoro: https://arxiv.org/abs/2608.22932 — 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 2026-10-18T10:08:38.287Z dalla pagina https://arxiv.org/catchup/eess.SP/2026-08-25?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).