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

arXiv: https://arxiv.org/abs/2610.03741 · HTML: https://mlsanity.com/paper/2610.03741/

## 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.

## Placed in

- Mixture-of-experts / [Efficient MoE training and inference systems](https://mlsanity.com/moe/efficient-moe-training-and-inference-systems/)

## Builds on (most cited first, 22 of 22)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Outrageously Large Neural Networks: The Sparsely-Gated Mixture-of-Experts Layer](https://mlsanity.com/paper/1701.06538/) (2017)
- [Switch Transformers: Scaling to Trillion Parameter Models with Simple and Efficient Sparsity](https://mlsanity.com/paper/2101.03961/) (2021)
- [GShard: Scaling Giant Models with Conditional Computation and Automatic Sharding](https://mlsanity.com/paper/2006.16668/) (2020)
- [Federated Learning via Over-the-Air Computation](https://mlsanity.com/paper/1812.11750/) (2018)
- [Broadband Analog Aggregation for Low-Latency Federated Edge Learning (Extended Version)](https://mlsanity.com/paper/1812.11494/) (2018)
- [Federated Learning over Wireless Fading Channels](https://mlsanity.com/paper/1907.09769/) (2019)
- [Distributed Learning in Wireless Networks: Recent Progress and Future Challenges](https://mlsanity.com/paper/2104.02151/) (2021)
- [One-Bit Over-the-Air Aggregation for Communication-Efficient Federated Edge Learning: Design and Convergence Analysis](https://mlsanity.com/paper/2001.05713/) (2020)
- [Generative AI Agents with Large Language Model for Satellite Networks via a Mixture of Experts Transmission](https://mlsanity.com/paper/2404.09134/) (2024)
- [Federated Edge Learning with Misaligned Over-The-Air Computation](https://mlsanity.com/paper/2102.13604/) (2021)
- [Learning Rate Optimization for Federated Learning Exploiting Over-the-air Computation](https://mlsanity.com/paper/2102.02946/) (2021)
- [Over-the-Air Computation Systems: Optimization, Analysis and Scaling Laws](https://mlsanity.com/paper/1909.00329/) (2019)
- [Task-Oriented Over-the-Air Computation for Multi-Device Edge AI](https://mlsanity.com/paper/2211.01255/) (2022)
- [Hybrid Beamforming for Massive MIMO Over-the-Air Computation](https://mlsanity.com/paper/2006.04560/) (2020)
- [Over-the-Air Computation via Reconfigurable Intelligent Surface](https://mlsanity.com/paper/2105.05113/) (2021)
- [Massive Digital Over-the-Air Computation for Communication-Efficient Federated Edge Learning](https://mlsanity.com/paper/2405.15969/) (2024)
- [The Larger the Merrier? Efficient Large AI Model Inference in Wireless Edge Networks](https://mlsanity.com/paper/2505.09214/) (2025)
- [Over-the-Air Computation via Intelligent Reflecting Surfaces](https://mlsanity.com/paper/1904.12475/) (2019)
- [Bayesian Over-The-Air Computation](https://mlsanity.com/paper/2109.03780/) (2021)
- [OTFS-based Robust MMSE Precoding Design in Over-the-air Computation](https://mlsanity.com/paper/2307.01525/) (2023)
- [SplitMAC: Wireless Split Learning over Multiple Access Channels](https://mlsanity.com/paper/2311.02405/) (2023)

## 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.