# Decoding the Skew: Distribution-Aware MoE Inference with Adaptive Kernel Dispatch

arXiv: https://arxiv.org/abs/2607.23099 · HTML: https://mlsanity.com/paper/2607.23099/

## 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 19 of its 27 references that are in the map.
- First posted on arXiv 25 Jul 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, 19 of 19)

- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [Qwen3 Technical Report](https://mlsanity.com/paper/2505.09388/) (2025)
- [Measuring Massive Multitask Language Understanding](https://mlsanity.com/paper/2009.03300/) (2020)
- [Efficient Memory Management for Large Language Model Serving with PagedAttention](https://mlsanity.com/paper/2309.06180/) (2023)
- [Switch Transformers: Scaling to Trillion Parameter Models with Simple and Efficient Sparsity](https://mlsanity.com/paper/2101.03961/) (2021)
- [Mixtral of Experts](https://mlsanity.com/paper/2401.04088/) (2024)
- [SGLang: Efficient Execution of Structured Language Model Programs](https://mlsanity.com/paper/2312.07104/) (2023)
- [Kimi K2: Open Agentic Intelligence](https://mlsanity.com/paper/2507.20534/) (2025)
- [CodeGeeX: A Pre-Trained Model for Code Generation with Multilingual Benchmarking on HumanEval-X](https://mlsanity.com/paper/2303.17568/) (2023)
- [FlashInfer: Efficient and Customizable Attention Engine for LLM Inference Serving](https://mlsanity.com/paper/2501.01005/) (2025)
- [MegaBlocks: Efficient Sparse Training with Mixture-of-Experts](https://mlsanity.com/paper/2211.15841/) (2022)
- [Accelerating Distributed MoE Training and Inference with Lina](https://mlsanity.com/paper/2210.17223/) (2022)
- [MoE-Lightning: High-Throughput MoE Inference on Memory-constrained GPUs](https://mlsanity.com/paper/2411.11217/) (2024)
- [EdgeMoE: Empowering Sparse Large Language Models on Mobile Devices](https://mlsanity.com/paper/2308.14352/) (2023)
- [SonicMoE: Accelerating MoE with IO and Tile-aware Optimizations](https://mlsanity.com/paper/2512.14080/) (2025)
- [Turn Waste into Worth: Rectifying Top-k Router of MoE](https://mlsanity.com/paper/2402.12399/) (2024)
- [Patterns behind Chaos: Forecasting Data Movement for Efficient Large-Scale MoE LLM Inference](https://mlsanity.com/paper/2510.05497/) (2025)
- [X-MoE: Enabling Scalable Training for Emerging Mixture-of-Experts Architectures on HPC Platforms](https://mlsanity.com/paper/2508.13337/) (2025)
- [Efficient CPU-GPU Collaborative Inference for MoE-based LLMs on Memory-Limited Systems](https://mlsanity.com/paper/2512.16473/) (2025)

## Abstract (arXiv)

Mixture-of-Experts (MoE) inference consists of sparse expert GEMMs whose shapes vary with the runtime routing distribution. Existing serving systems typically select fused-MoE kernels using static token-count buckets, ignoring the per-expert routing distribution that determines tile padding, memory reuse, and kernel efficiency. We introduce a distribution-aware framework for modeling and benchmarking MoE inference. The framework combines the compact Effective Experts metric with a Dirichlet-based reverse-modeling procedure that generates controllable routing distributions for systematic hardware studies. Using it, we show that the best fused-MoE kernel changes with routing skew and token count. We further present DA-MoE, a GPU-resident kernel-dispatch runtime for NVIDIA GPUs that matches the live routing histogram to offline-tuned distributions and selects a near-optimal fused-MoE kernel without CPU--GPU synchronization. On HumanEval-X serving traces, DA-MoE improves geomean fused-MoE latency by 1.16X on DeepSeek-V3 and 1.29X on Kimi K2, with peak speedups of 1.40X and 1.56X.