SiftMoE: Similarity-Aware Energy-Efficient Expert Selection for Wireless Distributed MoE Inference

[2603.23888] SiftMoE: Similarity-Aware Energy-Efficient Expert Selection for Wireless Distributed MoE Inference Skip to main content Search Submit Donate Log in Computer Science > Information Theory arXiv:2603.23888 (cs) [Submitted on 25 Mar 2026 ( v1 ), last revised 25 Aug 2026 (this version, v2)] Title: SiftMoE: Similarity-Aware Energy-Efficient Expert Selection for Wireless Distributed MoE Inference Authors: Qian Chen , Xianhao Chen , Kaibin Huang View PDF HTML (experimental) Abstract: Mixture-of-Experts (MoE) architectures leverage sparse activation to enhance the scalability of large language models (LLMs), making them suitable for deployment in resource-constrained edge networks. However, the sheer number of experts often exceeds the memory capacity of individual edge nodes, necessitating wireless distributed MoE (WIDE) inference where experts are spread across multiple edge nodes. In this context, expert selection directly affects communication costs. Motivated by the similarity of experts, we propose SiftMoE, which judiciously selects or skips experts to strike a tradeoff between communication costs and inference accuracy. Specifically, we first establish theoretical bounds on the accuracy degradation resulting from expert replacement or skipping. Based on the bounds, we formulate an energy minimization problem for expert selection in WIDE inference subject to latency and accuracy constraints. In particular, for slow-fading channels, we derive optimal expert selection policies for both single-token decoding and multi-token prefilling. For fast-fading channels, we further extend our scheme to cope with rapidly varying channel conditions. Simulation results demonstrate that SiftMoE significantly reduces energy consumption while maintaining inference accuracy compared with conventional Top-K routing in WIDE systems. Comments: 17 pages, 12 figures Subjects: Information Theory (cs.IT) ; Networking and Internet Architecture (cs.NI) Cite as: arXiv:2603.23888 [cs.IT] (or arXiv:2603.23888v2 [cs.IT] for this version) https://doi.org/10.48550/arXiv.2603.23888 Focus to learn more arXiv-issued DOI via DataCite Submission history From: Qian Chen [ view email ] [v1] Wed, 25 Mar 2026 03:23:51 UTC (193 KB) [v2] Tue, 25 Aug 2026 02:55:45 UTC (798 KB) Full-text links: Access Paper: View PDF HTML (experimental) TeX Source view license Additional Features Audio Summary Current browse context: cs.IT < prev | next > new | recent | 2026-03 Change to browse by: cs cs.NI math math.IT References & Citations SciX Google Scholar Semantic Scholar export BibTeX citation Bookmark Bibliographic Tools Bibliographic and Citation Tools Bibliographic Explorer Toggle Bibliographic Explorer ( What is the Explorer? ) Connected Papers Toggle Connected Papers ( What is Connected Papers? ) Litmaps Toggle Litmaps ( What is Litmaps? ) scite.ai Toggle scite Smart Citations ( What are Smart Citations? ) Code, Data, Media Code, Data and Media Associated with this Article alphaXiv Toggle alphaXiv ( What is alphaXiv? ) Links to Code Toggle CatalyzeX Code Finder for Papers ( What is CatalyzeX? ) DagsHub Toggle DagsHub ( What is DagsHub? ) GotitPub Toggle Gotit.pub ( What is GotitPub? ) Huggingface Toggle Hugging Face ( What is Huggingface? ) ScienceCast Toggle ScienceCast ( What is ScienceCast? ) Demos Demos Replicate Toggle Replicate ( What is Replicate? ) Spaces Toggle Hugging Face Spaces ( What is Spaces? ) Spaces Toggle TXYZ.AI ( What is TXYZ.AI? ) Related Papers Recommenders and Search Tools Link to Influence Flower Influence Flower ( What are Influence Flowers? ) Core recommender toggle CORE Recommender ( What is CORE? ) Author Venue Institution Topic About arXivLabs arXivLabs: experimental projects with community collaborators arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs . Which authors of this paper are endorsers? | Disable MathJax ( What is MathJax? ) We gratefully acknowledge support from our major funders , member institutions , and all contributors. About · Help · Contact · Subscribe · Copyright · Privacy · Accessibility · Operational Status (opens in new tab)

Related documents

• SiftMoE: Similarity-Aware Energy-Efficient Expert Selection for Wireless Distributed MoE Inference#670534
• Redundancy Meets Synergy: Dependency-aware Expert Selection for MoE via Submodular Optimization#1160986
• Redundancy Meets Synergy: Dependency-aware Expert Selection for MoE via Submodular Optimization#1155493
• Prediction-powered Inference by Mixture of Experts#146949
• Training-Free Halving of Activated Experts in Fine-Grained Mixture-of-Experts Models#927851