mlsanityData to 14 Oct 2026

Logit-Aware MIMO AirComp for Distributed Mixture-of-Experts LLM Inference over Wireless Edge Networks

arXiv 2610.03741 ↗First posted 22 Sep 2026cs.IT cs.AI cs.LGcited by 0

On the map: Mixture-of-experts → Efficient MoE training and inference systems also fits: Expert routing strategies

Abstract (arXiv)

Distributed mixture-of-experts (MoE) inference is a promising architecture for deploying large language models (LLMs) at wireless edge networks because sparse experts can be placed across coordinated base stations (BSs), while the anchor node and user equipment (UE) can offload LLM inference tasks to BSs. The communication bottleneck is the MoE aggregation, where the anchor BS must recover a weighted sum of selected expert outputs before each decoding step. Over-the-air computation (AirComp) is well matched to this operation because the wireless multiple-access channel naturally superposes simultaneous transmissions. However, conventional AirComp minimizes communication distortion, whereas MoE aggregation errors have unequal impact on LLM outputs. We propose a logit-aware MIMO AirComp framework that estimates local logit sensitivity as block-level weights and jointly optimizes receive combiners and BS precoders under per-BS power constraints. We also develop an alternating algorithm that protects decoding decisions from aggregation perturbations. Using OpenCompass, we evaluate Qwen3-30B-A3B-Instruct-2507-FP8 on GSM8K and ARC-Challenge. At 30 dB aggregation SNR, perturbed Qwen3 retains 99.3% of clean GSM8K accuracy and 94.8% of clean ARC-Challenge accuracy. In a direct ARC-Challenge closed-loop audit over 295 examples, SW-AirComp achieves 43.73% accuracy at 30 dB and 26.78% at 20 dB, outperforming unweighted, matched-filter, and zero-forcing AirComp. In the wireless simulator, SW-AirComp reduces decision-relevant aggregation distortion; at 20 dB, its final weighted-sum mean-squared error is 53.5% lower than unweighted AirComp under the same channels, samples, and sensitivity weights. Logit-RMSE and logit-gap diagnostics further show that the gain comes from reducing output-logit perturbation and lowering the risk of top-token changes.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in Efficient MoE training and inference systems, one of 12 directions in Mixture-of-experts. Builds on 22 of its 33 references that are in the map. First posted on arXiv 22 Sep 2026.

Builds on

22 of 33 references.

References · 22

  1. 2017Attention Is All You Needcited by 60,361
  2. 2017Outrageously Large Neural Networks: The Sparsely-Gated Mixture-of-Experts Layercited by 3,509
  3. 2021Switch Transformers: Scaling to Trillion Parameter Models with Simple and Efficient Sparsitycited by 3,073
  4. 2020GShard: Scaling Giant Models with Conditional Computation and Automatic Shardingcited by 1,631
  5. 2018Federated Learning via Over-the-Air Computationcited by 219
  6. 2018Broadband Analog Aggregation for Low-Latency Federated Edge Learning (Extended Version)cited by 191
  7. 2019Federated Learning over Wireless Fading Channelscited by 183
  8. 2021Distributed Learning in Wireless Networks: Recent Progress and Future Challengescited by 129
  9. 2020One-Bit Over-the-Air Aggregation for Communication-Efficient Federated Edge Learning: Design and Convergence Analysiscited by 112
  10. 2024Generative AI Agents with Large Language Model for Satellite Networks via a Mixture of Experts Transmissioncited by 35
  11. 2021Federated Edge Learning with Misaligned Over-The-Air Computationcited by 30
  12. 2021Learning Rate Optimization for Federated Learning Exploiting Over-the-air Computationcited by 29
10 more
  1. 2019Over-the-Air Computation Systems: Optimization, Analysis and Scaling Lawscited by 27
  2. 2022Task-Oriented Over-the-Air Computation for Multi-Device Edge AIcited by 14
  3. 2020Hybrid Beamforming for Massive MIMO Over-the-Air Computationcited by 13
  4. 2021Over-the-Air Computation via Reconfigurable Intelligent Surfacecited by 13
  5. 2024Massive Digital Over-the-Air Computation for Communication-Efficient Federated Edge Learningcited by 13
  6. 2025The Larger the Merrier? Efficient Large AI Model Inference in Wireless Edge Networkscited by 7
  7. 2019Over-the-Air Computation via Intelligent Reflecting Surfacescited by 5
  8. 2021Bayesian Over-The-Air Computationcited by 5
  9. 2023OTFS-based Robust MMSE Precoding Design in Over-the-air Computationcited by 2
  10. 2023SplitMAC: Wireless Split Learning over Multiple Access Channelscited by 2

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026AIR-LLM: Broadcasting AI Weights over Radio for Memory-Free Edge LLM Inference via RF Computing
  2. 2025Mixture-of-Experts for Distributed Edge Computing with Channel-Aware Gating Function
  3. 2024Adaptive Layer Splitting for Wireless LLM Inference in Edge Computing: A Model-Based Reinforcement Learning Approach
  4. 2024Collaborative Edge AI Inference over Cloud-RAN
  5. 2025Over-the-Air Federated Learning: Rethinking Edge AI Through Signal Processing
  6. 2024Private Collaborative Edge Inference via Over-the-Air Computation
  7. 2024WDMoE: Wireless Distributed Large Language Models with Mixture of Experts
  8. 2026EdgeXpert: An Edge Device for Memory-Efficient LLM Inference with Mixture-of-Experts and Speculative Decoding