Mixture-of-Experts Inference

Follow this topic Ask about this topic

Momentum

16 papers in the last four weeks, up 167% on the four weeks before. 0.2% of all new papers.

Jul 13Week of Sep 28

Latest papers 109

Jun 3, 2026cs.LG

SHAPE: Coalition-Aware Expert Pruning for Sparse Mixture-of-Experts LLMs

Sparse Mixture-of-Experts (MoE) large language models achieve strong quality with low per-token compute, yet their deployment is often limited by the memory wall: the full expert pool must remain resident to support token-dependent routing. Expert pruning is a direct remedy, but prior criteria typically score experts independently and overlook that MoE inference is inherently \emph{coalitional}, where outputs arise from routed top-

k

expert combinations. We propose \textbf{SHAPE}, a task-driven pruning framework that explicitly models \emph{intra-layer} expert cooperation. SHAPE formulates routing traces on a small calibration set as an empirical cooperative game and assigns interaction-aware expert values via a Shapley-style attribution over observed top-

k

coalitions, enabling the identification of experts that are essential for high-utility collaborations rather than merely frequent. To preserve MoE topology under a global pruning budget, SHAPE further introduces a \emph{quality-coverage} selection rule that retains, in each layer, the minimal expert subset covering an

α

fraction of non-negative Shapley mass, while using bisection to match a target keep rate. Experiments on three modern MoE backbones (Qwen3-30B-A3B, GPT-OSS-20B, and DeepSeek-V2-Lite) across diverse benchmarks show that SHAPE consistently improves robustness over global and layer-wise pruning variants, maintaining competitive accuracy under 20% and 40% expert pruning without additional training and delivering clear reductions in peak GPU memory footprint. The open-source code is available at https://github.com/Alizen-1009/Shapley-Moe.

Yuhao Zhang

Beihang University

Mixture-of-Experts Pruning LLM Pruning Sparse Mixture-of-Experts Mixture-of-Experts Inference Mixture-of-Experts Models

Jun 2, 2026cs.DC

UltraEP: Unleash MoE Training and Inference on Rack-Scale Nodes with Near-Optimal Load Balancing

Large-scale expert parallelism (EP) is becoming pivotal for training and serving frontier MoE models, but it also amplifies device-level expert load imbalance into compute stragglers, token all-to-all bottlenecks, and activation-memory spikes. Existing balancers redistribute experts periodically based on historical load, which becomes unreliable for production deployments with non-stationary load patterns. We present UltraEP, the first exact-load, real-time balancer for large-EP MoE training and serving prefill on rack-scale nodes (RSNs). Leveraging the extended scale-up connectivity among dozens of GPUs within RSNs, UltraEP rebalances every microbatch and layer on critical paths, which requires nontrivial co-design of plan solving and expert replication communication to minimize exposed overhead. To this end, UltraEP eagerly reacts to post-gating load with an efficient quota-driven planner, and executes the resulting irregular expert-state transfers with RSN-native persistent tile streaming and relay-based fan-out mitigation. We evaluate UltraEP in a multi-RSN deployment of up to 256 GPUs, using cutting-edge MoE models from 106B to 671B parameters. Averaged across training and serving, UltraEP achieves 94.3% of the force-balanced ideal throughput, delivering 1.49

\times

improvement over no-balancing, while reducing the final inter-rank imbalance from 1.30

-

4.01 to 1.01

-

1.04.

Xinming Wei, Chao Jin, Tuo Dai +10

School of Computer Science, Peking University · Xiaohongshu Inc. · Shanghai AI Laboratory +1

Expert Parallelism Expert Load Balancing High-Performance Computing Mixture-of-Experts Inference Distributed Training Mixture-of-Experts Models+1

May 31, 2026cs.LG

Beyond Task-Agnostic: Task-Aware Grouping for Communication-Efficient Multi-Task MoE Inference

Sparsely activated Mixture-of-Experts (MoE) models scale capacity via conditional computation, but distributed inference suffers from cross-GPU expert communication and routing-induced load imbalance. Existing placement methods reduce this cost by co-locating frequently co-activated experts; however, they derive a single deployment plan from globally aggregated routing traces, thereby averaging away the heterogeneous, task-specific co-activation patterns that actually drive communication in multi-task serving. We observe that expert co-activation is strongly task-conditioned: pairs tightly coupled in one task family are often uncorrelated in another, so effective deployment should group experts by task-aware co-activation rather than by a task-agnostic average. Based on this insight, we propose \emph{Task-Aware Coactivation Grouping} (TACG), a deployment-time framework that uses family-specific dispatch and co-activation traces to derive per-expert task-family preferences, reweights the co-activation graph so that intra-family locality dominates grouping, and assigns each expert to a primary GPU under exact capacity constraints. To keep the static placement robust under online workload skew, we further introduce \emph{Generic Expert Shared Replication} (GESR), a lightweight companion that identifies generic experts with consistently central co-activation profiles, replicates them across a small set of secondary GPUs, and applies locality- and load-aware selection at serving time. Experiments on three representative open-source MoE models demonstrate that our framework reduces the average communication cost by 31.39% over the baseline, while preserving an average Jain fairness index of 0.9975. This advantage persists even under severe distribution shifts in the inference data, consistently outperforming strong baselines.

Zhiyao Xu, Aoxue Liu, Zhanjie Ding +3

Tsinghua Shenzhen International Graduate School, Shenzhen, China · Pengcheng Laboratory, Shenzhen, China

Mixture-of-Experts Inference Mixture-of-Experts Models Distributed Inference

May 30, 2026cs.DC

ViBE: Co-Optimizing Workload Skew and Hardware Variability for MoE Serving

In distributed Mixture-of-Experts (MoE) inference, input-dependent token routing interacts with GPU performance variability to create persistent stragglers under synchronized execution, where the slowest GPU determines layer latency. This performance variability is inherent to modern accelerators: manufacturing variation, power limits, and thermal conditions introduce measurable execution-time differences across nominally identical GPUs. The core challenge is that MoE execution-time imbalance arises from the interaction of workload skew and hardware asymmetry. Token routing produces uneven and layer-varying expert loads, while GPU throughput depends on device-specific operating characteristics and workload intensity. Prior work mitigates routing skew but assumes homogeneous hardware, optimizing token balance rather than execution latency. As a result, even balanced token assignments can leave hardware-induced stragglers unaddressed. Thus, we propose Variability-Informed Binning of Experts (ViBE), a hardware-aware expert placement framework that minimizes execution-time imbalance across GPUs. ViBE combines per-GPU performance modeling with expert activation profiling to assign high-load experts to faster devices and low-load experts to slower ones, reducing layer-level stragglers without modifying model semantics or hardware. Because both workload characteristics and effective GPU throughput can shift across serving conditions, ViBE supports lightweight recalibration under workload/performance drift to refresh its routing and performance estimates when needed. Results show that ViBE consistently reduces execution-time imbalance and improves SLO attainment by 14%, while lowering P90 TTFT by up to 45%. We further show that the impact of hardware variability increases at scale, making variability-aware placement important for efficient, high-utilization LLM serving.

Seokjin Go, Marko Scrbak, Ephrem Wu +2

Georgia Institute of Technology Atlanta, GA, USA · Advanced Micro Devices, Inc. Austin, TX, USA · Advanced Micro Devices, Inc. Santa Clara, CA, USA +1

Expert Load Balancing Mixture-of-Experts Inference LLM Inference Acceleration

May 28, 2026cs.LG

Composing Non-Conjugate Factor Graphs with Closed-Form Variational Inference

Stacking probabilistic building blocks into deeper architectures typically breaks closed-form inference. We show that closed-form inference can be preserved. We identify five factor-graph primitives: a bilinear factor, an exponential link, a Gamma prior, a Gaussian likelihood, and an equality node, and prove that any model composed from them admits closed-form variational message passing. The construction works because each primitive preserves a small set of message families: under mean-field factorization, messages on Gaussian variables remain Gaussian and messages on precision variables remain Gamma, while the only non-conjugate interface, the exponential link, remains tractable through the Gaussian moment-generating function and the sufficient statistics of the Gamma family. We demonstrate composition at increasing depth, from static ensembles through input-dependent gating to split-branch routing, and show that stacking routing layers encodes arbitrary decision trees, establishing universal function approximation with closed-form inference. Applied to ensemble time-series forecasting, the framework yields a Bayesian mixture of experts in which gating functions are inferred rather than learned, providing calibrated uncertainty over expert selection across five benchmark datasets.

Mykola Lukashchuk, Kyrylo Yemets, Wouter M. Kouw +4

Eindhoven University of Technology, the Netherlands · Lviv Polytechnic National University, Lviv, Ukraine · Lazy Dynamics, Utrecht, the Netherlands +1

Factor Graph Optimization Mixture-of-Experts Inference Probabilistic Graphical Models Variational Inference

May 27, 2026cs.LG

How Far Can Disaggregation Go? A Design-Space Exploration of Attention-FFN Disaggregation for Efficient MoE LLM Serving

Modern large language model (LLM) inference has progressively disaggregated to keep pace with growing model sizes and tight TTFT and TPOT service-level objectives: from chunked-prefill aggregation, to prefill-decode (P/D) disaggregation, and most recently to operator-level Attention-FFN Disaggregation (AFD). This trend is especially important for mixture-of-experts (MoE) models, where memory-bound attention, compute-intensive expert FFNs, and MoE dispatch/combine communication create distinct resource demands. AFD further exposes this heterogeneity by placing attention and MoE-FFN execution on separate GPU groups. Each level of disaggregation deepens the scheduling design space across workload characteristics, resource allocation, and interconnect topology, raising the central question: when does each level actually pay off? We systematically characterize this trade-off for MoE inference across realistic workloads spanning input/output sequence lengths, prefix-KV reuse, and per-user latency constraints. Using chunked-prefill and P/D disaggregation as baselines, we study the benefits and limits of AFD at scale through a framework that fuses on-device kernel measurements with high-fidelity network simulation. Under strict TTFT/TPOT SLOs, AFD sustains around 4k tokens/s of system throughput on DeepSeek-V3.2 across chat, coding, and agentic-coding workloads, where non-AFD deployments are infeasible. We distill concrete takeaways for jointly optimizing throughput and interactivity, including how to partition attention and FFN across GPUs as a function of workload and model architecture, providing design principles for current rack- and cluster-scale deployments as well as future disaggregated AI infrastructure.

Hanjiang Wu, Abhimanyu Rajeshkumar Bambhaniya, Sarbartha Banerjee +9

1Georgia Institute of Technology · 5Infravana · 2Intel +2

LLM Serving LLM Inference Disaggregated LLM Serving Mixture-of-Experts Inference Distributed Inference

May 26, 2026cs.LG

MobileMoE: Scaling On-Device Mixture of Experts

Mixture-of-Experts (MoE) has become the de facto architecture for hundred-billion-parameter language models, yet its advantages at sub-billion scales for on-device deployment remain largely unexplored. To close this gap, we present MobileMoE, a family of on-device MoE language models with sub-billion active parameters (0.3-0.9B active and 1.3-5.3B total) that establish a new Pareto frontier for on-device LLMs. We first formulate an on-device MoE scaling law that jointly optimizes MoE architecture under mobile memory and compute constraints, identifying an on-device sweet spot - moderate sparsity with fine-grained and shared experts - that is simultaneously memory and compute-optimal. Building on the derived architectures, we train MobileMoE with a four-stage recipe covering pre-training, mid-training, instruction fine-tuning, and quantization-aware training, all on open-source datasets. Across 14 benchmarks, MobileMoE matches or exceeds leading on-device dense LLMs with 2-4

\times

fewer inference FLOPs, and matches or surpasses the state-of-the-art MoE OLMoE-1B-7B with up to 60% fewer parameters. To bridge the last mile to mobile deployment, we provide the first efficient MoE inference on commodity smartphones with comprehensive on-device profiling. At comparable INT4 weight memory, MobileMoE-S delivers

1.8

-

3.8\times

faster prefill and

2.2

-

3.4\times

faster decode than the dense baseline MobileLLM-Pro.

Yanbei Chen, Hanxian Huang, Ernie Chang +5

Meta AI

Mixture-of-Experts Language Models On-Device Language Model Inference Mixture-of-Experts Inference Efficient Language Model Inference AI Scaling Laws

May 26, 2026cs.LG

ReMoE: Boosting Expert Reuse through Router Fine-Tuning in Memory-Constrained MoE LLM Inference

Fine-grained Mixture-of-Experts (MoE) models sparsely activate only a subset of experts per token, reducing activated computation while maintaining high model capacity. However, in memory-constrained inference scenarios, only a small set of experts can be cached. Experts not in the cache must be fetched from slow external storage (e.g., UFS), leading to frequent evictions and substantial I/O overhead. We propose ReMoE, a router fine-tuning framework designed to boost token-wise expert reuse. ReMoE biases the router toward recently selected experts, producing temporally stable routing that better matches cache locality constraints. By increasing short-horizon expert reuse, ReMoE reduces expert fetches from storage without adding inference-time computation. Experiments on DeepSeek and Qwen models show that ReMoE improves expert reuse by 26% while maintaining downstream task performance. Real-system evaluations further confirm these benefits, improving output throughput by 8.4% under vLLM GPU-CPU expert offloading and reducing TPOT by 43.6-49.8% under llama.cpp on Jetson Orin NX, corresponding to a 1.77-1.99

\times

decode speedup across diverse workloads. Checkpoints and usage instructions are available at https://github.com/BUAA-OSCAR/ReMoE.

Xiongwei Zhu, Xiaojian Liao, Tianyang Jiang +3

School of Computer Science and Engineering, Beihang University, Beijing 100191, China · Huawei Technologies Ltd.

LLM Inference Efficiency Expert Offloading Fine-Tuning Mixture of Experts Memory-Efficient Optimization Mixture-of-Experts Inference+1

May 26, 2026cs.LG

Dense2MoE: Pushing the Pareto Frontier of On-Device LLMs via Unified Pruning and Upcycling

The Mixture of Experts MoE architecture is highly promising for resource constrained on device deployments yet training these models from scratch incurs prohibitive costs Current methods attempt to alleviate this by upcycling dense models into MoEs however they often introduce parameter redundancy that degrades inference efficiency Alternatively standard layer pruning mitigates redundancy but inevitably compromises model accuracy To resolve this dilemma we propose Dense2MoE a novel framework that unifies pruning and upcycling through Layer Fusion UpCycling LF UC Guided by hardware Roofline theory Dense2MoE systematically overcomes the inference memory wall by pruning bandwidth heavy attention modules from redundant layers while repurposing their Multi Layer Perceptrons MLPs into MoE experts This structural innovation preserves the models core capabilities and strictly limits active parameters via selective token routing With a modest continual pre training budget Dense2MoE efficiently converts publicly available dense LLMs into on device ready MoE models Extensive experiments demonstrate that Dense2MoE significantly advances the Pareto frontier for on device inference latency versus model accuracy outperforming dense baselines state of the art compression and standard upcycling methods

Fengfa Li, Hongjin Ji, Yifeng Ding +2

Li Auto · The Chinese University of Hong Kong, Shenzhen

Mixture-of-Experts Pruning LLM Pruning On-Device Language Model Inference LLM Compression Mixture-of-Experts Inference

May 22, 2026cs.CV

DepthAgent: Towards Better Universal Depth Estimation via Sample-wise Expert Selection

Monocular metric depth estimation has achieved strong progress with large-scale training and universal-camera modeling, yet robust deployment across diverse camera settings, such as perspective, fisheye, and panoramic images, remains challenging. Existing methods typically rely on a single depth estimator, overlooking that different models encode different camera assumptions and perform best under different input domains. In this paper, we show that depth experts exhibit strong sample-wise complementarity: model preference is highly correlated with camera geometry, and multi-model fusion brings the largest gains on difficult samples where individual experts are unreliable. Motivated by these observations, we propose \textbf{\ours}, a vision-language agent for adaptive monocular depth estimation. DepthAgent treats existing depth models as frozen tools and learns to analyze scene and camera cues, invoke suitable experts through multi-turn tool utilization, and select or fuse their predictions for each input. To optimize such discrete decision-making toward dense geometric quality, we design a multi-reward reinforcement fine-tuning scheme that jointly encourages valid tool execution, camera/scene analysis, expert-selection quality, and inference efficiency. Extensive experiments across perspective, fisheye, and panoramic benchmarks show that \ours consistently outperforms individual experts, fixed model fusion, and different selection strategies, with strong improvements on challenging samples, highlighting the critical role of expert selection and fusion. The code and model will be released upon publication.

Jie Zhu, Girish Chandar Ganesan, Xiaoming Liu

Michigan State University · University of North Carolina at Chapel Hill

Metric Depth Estimation Monocular Depth Estimation Mixture-of-Experts Inference Vision-Language Agents

May 20, 2026cs.DC

Diagnosing Overhead in Dispatch Operations: Cross-architecture Observatory

AlltoAll dispatch is the dominant bottleneck of MoE expert parallelism, and the interconnect community has responded with four families of mitigations: predictive sample placement, adaptive expert relayout, hierarchical collectives, and EP-aware topology. All four rest on two assumptions about the workload. The first is that routing imbalance is correctable by the system layer. The second is that the mock-token benchmarks evaluating them faithfully represent production routing. We introduce DODOCO to test both assumptions. We instrument five MoE checkpoints spanning five sequence-mixer designs (DeepSeek-V2-Lite MLA, DeepSeek-MoE-16B MHA, Qwen3-30B GQA, Nemotron-30B Mamba-2, Qwen3.5-35B GDN) under a 5 by 6 grid of data conditions plus a matched EP scan from 4 to 32 ranks on H100s; both assumptions fail. Scaling EP changes the per-expert max/mean token ratio by at most 5% within every architecture's measurable range: the straggler is intrinsic to the routing decision the model makes, not to how its experts land on ranks. Mock tokens overestimate routing Gini by up to a factor of 2.35 and fabricate a batch-size scaling trend that vanishes the moment real text replaces random IDs. A third pattern, unexpected, emerges from the same matrix: the five architectures cleave into two stable bands. MHA and Mamba-2 (data-resilient) drop to Gini 0.105 and 0.150 on wikitext. MLA and GDN (persistently concentrated) stay above 0.24 on every real-text condition and reach 0.29 to 0.38 on mock. GQA is the intermediate case. These bands, not the EP degree or the mock-data profile, are the right workload input to AlltoAll-aware interconnect and dispatch design.

Bole Ma, Jan Eitzinger, Harald Koestler +1

Expert Parallelism Expert Load Balancing Benchmark Design High-Performance Computing Mixture-of-Experts Inference

May 19, 2026cs.CL

TIDE: Efficient and Lossless MoE Diffusion LLM Inference with I/O-aware Expert Offload

Diffusion Large Language Models (dLLMs) have emerged as a competitive alternative to autoregressive (AR) models, offering better hardware utilization and bidirectional context through parallel block-level decoding. However, as dLLMs continue to scale up with mixture-of-experts (MoE) architectures, their deployment on resource-constrained devices remains an open challenge. Existing AR-based methods often incur either prohibitive I/O overhead or significant compute bottlenecks. In this work, we propose TIDE, a novel resource-efficient inference system that leverages the temporal stability of expert activations during the diffusion process within the block. Specifically, we leverage the temporal stability of expert activations during the diffusion process within the block and introduce an interval-based expert refresh strategy that updates the expert placement in an I/O-aware fashion. To ensure optimal performance, we formulate the inference scheduling as a mathematical programming problem, solving for the optimal interval that minimizes I/O traffic and CPU computation. Most importantly, TIDE is a lossless optimization that requires no model training, providing a "free lunch" acceleration for dLLM inference. In a single GPU-CPU system, we demonstrate that TIDE achieves up to 1.4

\times

and 1.5

\times

throughput improvements over prior baselines on LLaDA2.0-mini and LLaDA2.0-flash models, respectively.

Zhiben Chen, Youpeng Zhao, Yang Sui +2

University of Central Florida · Rice University

Expert Offloading LLM Inference Scheduling Mixture-of-Experts Inference Diffusion Language Model Inference Diffusion Model Inference Acceleration

May 19, 2026cs.DC

GEM: GPU-Variability-Aware Expert to GPU Mapping for MoE Systems

Mixture-of-Expert (MoE) models enable efficient inference by employing smaller experts and activating only a subset of them per token. MoE serving engines distribute experts across multiple GPUs and route tokens to appropriate GPUs at inference time based on experts activated. They process tokens in lock-step fashion, where tokens within a batch must finish processing before proceeding to the next layer. This synchronization barrier acts as a critical bottleneck because the performance of MoE models is limited by the straggler GPU that finishes last. Stragglers emerge when too many heavily used experts are placed on the same GPU or the slowest GPU. While prior works place experts that balance token loads across GPUs, they all overlook GPU variability and often place highly used experts on the slowest GPUs. We propose GEM, GPU-variability-aware Expert Mapping, a framework for GPU variability-aware expert to GPU mapping for MoE models. GEM exploits two insights. First, we must place experts such that each GPU receives non-uniform token loads based on their variability and they all finish processing a layer at about the same time. Our studies show that there are two types of experts: consistent that are used most of the time and temporal that are often used together for the remaining time. Our second insight is that we must place simultaneously used consistent and temporal experts on different GPUs and avoid placing them on slower GPUs to reduce slowdown. GEM gathers the variability profile of GPUs for each model and task and uses the token load distributions per task to map experts to GPUs. Our experiments show that GEM improves end-to-end latency by 7.9% on average and by up to 16.5% compared to the baseline.

Sourish Wawdhane, Avinash Kumar, Poulami Das

The University of Texas at Austin

Expert Load Balancing Mixture-of-Experts Inference Heterogeneous Computing Mixture-of-Experts Models

May 19, 2026cs.CV

FPED: A Functional-Network Prior-Guided Mixture-of-Experts Framework for Interpretable Brain Decoding

Visual image reconstruction from functional Magnetic Resonance Imaging (fMRI) is a fundamental task in brain decoding, providing a crucial pathway for understanding human perceptual mechanisms and developing advanced brain-computer interfaces (BCIs). However, most current methods simply flatten fMRI signals from localized visual cortices into one-dimensional (1D) vectors, mapping them directly into latent spaces such as that of Contrastive Language-Image Pre-training (CLIP). This paradigm not only disrupts the inherent network topology of the brain-leading to limited neuroscientific interpretability-but also overlooks the synergistic contributions of other distributed functional networks in processing high-level visual semantics. To address these limitations, we propose FPED, a Functional-Network Prior-Guided Mixture of Experts (MoE) framework for interpretable brain decoding. FPED explicitly models different functional brain networks as specialized experts and employs adaptive routing to capture their complementary contributions to visual semantic understanding. Unlike conventional homogeneous decoding paradigms, our framework incorporates neurobiologically grounded priors to enable structured and interpretable network-level representation learning. Experimental results demonstrate that FPED achieves highly competitive semantic reconstruction performance with only 0.68B parameters. The learned routing dynamics reveal biologically meaningful correspondence between functional brain networks and modality-specific semantic processing, providing transparent neuroscientific interpretability. This suggests that brain network-aware expert modeling is a promising direction for bridging neural decoding and biologically inspired artificial intelligence.

Yudan Ren, Pengcheng Shi, Zihan Ma +2

School of Electronic Information (School of Artificial Intelligence), Northwest University, Xi’an 710127, China

Neural Decoding Mixture-of-Experts Inference Neural Network Interpretability Neuroimaging

May 18, 2026cs.LG

Post-Trained MoE Can Skip Half Experts via Self-Distillation

Mixture-of-Experts (MoE) scales language models efficiently through sparse expert activation, and its dynamic variant further reduces computation by adjusting the activated experts in an input-dependent manner. Existing dynamic MoE methods usually rely on pre-training from scratch or task-specific adaptation, leaving the practical conversion of fully trained MoE underexplored. Enabling such adaptation would directly alleviate the inference costs by allowing easy tokens to bypass unnecessary expert during serving. This paper introduces Zero-Expert Self-Distillation Adaptation (ZEDA), a low-cost framework that transforms post-trained static MoE models into efficient dynamic ones. To stabilize this architectural conversion, ZEDA injects parameter-free zero-output experts into each MoE layer and adapts the augmented model through two-stage self-distillation, utilizing the original MoE as a frozen teacher and applying a group-level balancing loss. On Qwen3-30B-A3B and GLM-4.7-Flash across 11 benchmarks spanning math, code, and instruction following, ZEDA eliminates over 50% of expert FLOPs at marginal accuracy loss. It outperforms the strongest dynamic MoE baseline by 6.1 and 4.0 points on the two models, and delivers ~1.20

\times

end-to-end inference speedup.

Xingtai Lv, Li Sheng, Kaiyan Zhang +12

Tsinghua University · Shanghai AI Lab · WeChat AI +1

Mixture of Experts Mixture-of-Experts Inference LLM Inference Acceleration Knowledge Distillation

May 18, 2026cs.LG

CoX-MoE: Coalesced Expert Execution for High-Throughput MoE Inference with AMX-Enabled CPU-GPU Co-Execution

The Mixture-of-Experts (MoE) architecture improves computational efficiency via sparse expert activation, but throughput-oriented inference faces substantial GPU memory pressure due to a significant parameter size and intermediate data. Prior works attempt to mitigate this using expert offloading with micro-batching or by offloading computation to the CPU. However, the fragmented workload resulting from micro-batching degrades operational intensity, causing expert execution to become memory-bound. Meanwhile, CPU offloading is constrained by slow PCIe transfers and its limited applicability to attention computation in the decode stage. Consequently, these inefficiencies prevent effective system utilization, severely restricting the end-to-end throughput of MoE inference. To address these challenges, this paper proposes CoX-MoE, an Advanced Matrix Extensions (AMX)-enabled CPU-GPU collaborative system that comprehensively optimizes MoE inference by combining coalesced expert execution with strategic workload orchestration for higher throughput. CoX-MoE introduces (i) a coalescing-aware orchestration policy to jointly optimize resource allocation by adopting ordinary batch, instead of micro-batch, for expert computation and selective attention offloading, and (ii) a static expert-aware stratification scheme that pre-assigns frequently activated experts to the GPU, mitigating PCIe transfer overhead and balancing workload for the CPU and GPU during inference. Compared to state-of-the-art frameworks, CoX-MoE delivers significant gains, achieving up to 7.1x and 2.4x higher throughput than FlexGen and MoE-Lightning, respectively.

Muyoung Son, Yi Chen, Seungjae Yoo +2

KAIST · Daejeon, Republic of Korea

Expert Offloading GPU Acceleration Mixture-of-Experts Inference LLM Inference Acceleration Heterogeneous Computing

May 14, 2026cs.AI

BEAM: Binary Expert Activation Masking for Dynamic Routing in MoE

Mixture-of-Experts (MoE) architectures enhance the efficiency of large language models by activating only a subset of experts per token. However, standard MoE employs a fixed Top-K routing strategy, leading to redundant computation and suboptimal inference latency. Existing acceleration methods either require costly retraining with architectural changes or suffer from severe performance drop at high sparsity due to train-inference mismatch. To address these limitations, we propose BEAM (Binary Expert Activation Masking), a novel method that learns token-adaptive expert selection via trainable binary masks. With a straight-through estimator and an auxiliary regularization loss, BEAM induces dynamic expert sparsity through end-to-end training while maintaining model capability. We further implement an efficient custom CUDA kernel for BEAM, ensuring seamless integration with the vLLM inference framework. Experiments show that BEAM retains over 98% of the original model's performance while reducing MoE layer FLOPs by up to 85%, achieving up to 2.5

\times

faster decoding and 1.4

\times

higher throughput, demonstrating its effectiveness as a practical, plug-and-play solution for efficient MoE inference.

Juntong Wu, Jialiang Cheng, Qishen Yin +5

Taobao & Tmall Group of Alibaba · Shenzhen Graduate School, Peking University

Sparse Mixture-of-Experts Mixture-of-Experts Inference LLM Inference Acceleration LLM Routing

May 12, 2026cs.LG

ROMER: Expert Replacement and Router Calibration for Robust MoE LLMs on Analog Compute-in-Memory Systems

Large language models (LLMs) with mixture-of-experts (MoE) architectures achieve remarkable scalability by sparsely activating a subset of experts per token, yet their frequent expert switching creates memory bandwidth bottlenecks that compute-in-memory (CIM) architectures are well-suited to mitigate. However, analog CIM systems suffer from inherent hardware imperfections that perturb stored weights, and its negative impact on MoE-based LLMs in noisy CIM environments remains unexplored. In this work, we present the first systematic investigation of MoE-based LLMs under noise model calibrated with real chip measurements, revealing that hardware noise critically disrupts expert load balance and renders clean-trained routing decisions consistently suboptimal. Based on these findings, we propose ROMER, a post-training calibration framework that (1) replaces underactivated experts with high-frequency ones to restore load balance, and (2) recalibrates router logits via percentile-based normalization to stabilize routing under noise. Extensive experiments across multiple benchmarks demonstrate that ROMER achieves up to 58.6%, 58.8%, and 59.8% reduction in perplexity under real-chip noise conditions for DeepSeek-MoE, Qwen-MoE, and OLMoE, respectively, establishing its effectiveness and generalizability across diverse MoE architectures.

Wenyong Zhou, Yuannuo Feng, Yizhe Chen +6

The Department of Electrical and Computer Engineering, The University of Hong Kong, Hong Kong · The School of Integrated Circuit Science and Engineering, Beihang University, Beijing, China

Compute-in-Memory Mixture-of-Experts Language Models Mixture of Experts Mixture-of-Experts Inference Language Model Calibration Language Model Robustness

May 12, 2026cs.LG

Fast MoE Inference via Predictive Prefetching and Expert Replication

The Mixture of Experts (MoE) architecture has become a fundamental building block in state-of-the-art large language models (LLMs), improving domain-specific expertise in LLMs and scaling model capacity without proportionally increasing their computational overhead. However, MoE inference often suffers from suboptimal GPU utilization, load imbalance, and elevated latency arising from multiple tokens waiting on the same experts for their computation which arises from sparsity of expert activation. To address these challenges, we propose a dynamic expert replication strategy that predicts which experts are likely to be overloaded and replicates them for upcoming batches of tokens. The replicated experts process batch tokens concurrently across layers, which leads to improved parallelism, shorter GPU idle time, and significantly faster inference. Experimental evaluations conducted on large-scale MoE models, including Switch-base-128 and Switch-base-256, demonstrate that our method achieves near-complete GPU utilization (approx 100%), leading to upto 3x improvement in inference speed while preserving approximately 90-95% of the performance of baseline architectures

Ankit Jyothish, Ali Jannesari, Aishwarya Sarkar +1

Iowa State University Ames, USA

Expert Load Balancing GPU Acceleration Mixture of Experts Mixture-of-Experts Inference LLM Inference Acceleration

May 12, 2026cs.LG

TRACE: Temporal Routing with Autoregressive Cross-channel Experts for EEG Representation Learning

Learning transferable representations for electroencephalography (EEG) remains challenging because EEG signals are inherently multi-channel and non-stationary. Channels observed at the same time provide coupled measurements of neural activity, while the relevant temporal dynamics vary across contexts. This structure is poorly matched by architectures that apply uniform computation across time or route each channel patch independently. To this end, we propose TRACE, an autoregressive EEG pre-training framework that predicts future EEG patches from causal context while performing temporally adaptive and cross-channel coherent computation. At each temporal step, TRACE derives an expert routing decision from the causal cross-channel history and applies it jointly to all channels at that step. This preserves instantaneous cross-channel coherence while allowing different temporal regimes to activate different computation. Since routing is defined over the available channel set and causal temporal context, TRACE is compatible with heterogeneous pre-training across corpora with different channel counts, montages, sequence lengths, and recording domains. Across eight downstream EEG benchmarks, TRACE is evaluated in both settings: when downstream domains are seen only as unlabeled pre-training data and when downstream datasets are completely unseen during pre-training. It obtains the best results on several benchmarks while remaining competitive on motor imagery and clinical event classification tasks, with ablations supporting the importance of cross-channel temporal routing.

Fan Ma, Qier An, Peng Chen +6

Department of Biomedical Informatics and Data Science, Yale University

Representation Learning Self-Supervised Pre-Training Mixture-of-Experts Inference Predictive Representation Learning EEG Foundation Models

May 11, 2026cs.LG

DECO: Sparse Mixture-of-Experts with Dense-Comparable Performance on End-Side Devices

While Mixture-of-Experts (MoE) scales model capacity without proportionally increasing computation, its massive total parameter footprint creates significant storage and memory-access bottlenecks, which hinder efficient end-side deployment that simultaneously requires high performance, low computational cost, and small storage overhead. To achieve these properties, we present DECO, a sparse MoE architecture designed to match the performance of dense Transformers under identical total parameter budgets and training tokens. DECO utilizes the differentiable and flexible ReLU-based routing enhanced by learnable expert-wise scaling, which adaptively balances the contributions of routed and shared experts. Furthermore, we introduce NormSiLU, an activation function that normalizes inputs prior to SiLU operators, producing a more stable trend of routed-expert activation ratio and a higher intrinsic sparsity level. We also identify an empirical advantage in using non-gated MLP experts with ReLU-based routing, indicating the possibility of MoE architecture simplification. Experiments demonstrate that DECO, activating only 20% of routed experts, matches dense performance and outperforms established MoE baselines. Our specialized acceleration kernel delivers a 2.93

\times

speedup on Jetson AGX Orin compared with dense inference. Code and checkpoints are available at https://github.com/thunlp/DECO.

Chenyang Song, Weilin Zhao, Xu Han +3

Dept. of Comp. Sci. & Tech., Institute for AI, Tsinghua University, Beijing, China

Sparse Mixture-of-Experts Mixture-of-Experts Inference Expert Routing On-Device Inference

May 10, 2026cs.LG

TileQ: Efficient Low-Rank Quantization of Mixture-of-Experts with 2D Tiling

Mixture-of-Experts (MoE) models achieve remarkable performance by sparsely activating specialized experts, yet their massive parameters in experts pose significant challenges for deployment. While low-rank quantization offers a promising route to compress MoE models, existing methods still incur nonnegligible memory overhead and inference latency. To address these limitations, we propose \textsc{TileQ}, a fine-tuning-free post-training quantization (PTQ) method that employs 2D-tiling structured low-rank quantization to share low-rank factors across both input and output dimensions of MoE experts. Furthermore, we introduce an efficient inference technique for \textsc{TileQ} that fuses multiple low-rank expert computations into a single-pass operation, significantly improving hardware utilization. Experiments show that \textsc{TileQ} cuts down additional memory usage up to 10

\times

and reduces inference latency to

\sim

5% while preserving state-of-the-art accuracy.

Hongyaoxing Gu, Xinzhe Chen, Lijuan Hu +1

Institute of Software Chinese Academy of Sciences · University of Chinese Academy of Sciences

Efficient Inference Mixture-of-Experts Inference Mixture-of-Experts Quantization Mixture-of-Experts Models Post-Training Quantization Neural Network Quantization

May 9, 2026cs.LG

Uncovering Intra-expert Activation Sparsity for Efficient Mixture-of-Expert Model Execution

Mixture of Experts (MoE) architecture has become the standard for state-of-the-art large language models, owing to its computational efficiency through sparse expert activation. However, sparsity through finer expert granularity is becoming increasingly difficult to achieve due to fundamental training challenges such as expert collapse and load imbalance. In this work, we explore and leverage intra-expert activation sparsity as a complementary and underexplored dimension of sparsity in MoE models. Surprisingly, substantial intra-expert sparsity is readily available in existing pre-trained MoE models, without any modification to the activation function or model parameters, providing up to 90% sparsity within each expert without significant accuracy loss. We explore intra-expert activation sparsity across eight off-the-shelf MoE models ranging from 1B to 400B parameters, and extend the MoE execution pipeline of vLLM to leverage intra-expert activation sparsity by skipping the computations of inactive neurons, on top of its existing optimizations, achieving up to 2.5 times speedup in MoE layer execution and 1.2 times end-to-end speedup compared to the original dense vLLM baseline.

Jongseok Park, Sunga Kim, Zhenyu Gu +2

University of California, Berkeley · Work done during an internship at Advanced Micro Devices, Inc. · Advanced Micro Devices, Inc.

Activation Sparsity Mixture-of-Experts Inference LLM Inference Acceleration Mixture-of-Experts Models

May 7, 2026cs.CL

EMO: Pretraining Mixture of Experts for Emergent Modularity

Large language models are typically deployed as monolithic systems, requiring the full model even when applications need only a narrow subset of capabilities, e.g., code, math, or domain-specific knowledge. Mixture-of-Experts (MoEs) seemingly offer a potential alternative by activating only a subset of experts per input, but in practice, restricting inference to a subset of experts for a given domain leads to severe performance degradation. This limits their practicality in memory-constrained settings, especially as models grow larger and sparser. We introduce EMO, an MoE designed for modularity-the independent use and composition of expert subsets-without requiring human-defined priors. Our key idea is to encourage tokens from similar domains to rely on similar experts. Since tokens within a document often share a domain, EMO restricts them to select experts from a shared pool, while allowing different documents to use different pools. This simple constraint enables coherent expert groupings to emerge during pretraining using document boundaries alone. We pretrain a 1B-active, 14B-total EMO on 1T tokens. As a full model, it matches standard MoE performance. Crucially, it enables selective expert use: retaining only 25% (12.5%) of experts incurs just a 1% (3%) absolute drop, whereas standard MoEs break under the same setting. We further find that expert subsets in EMO specialize at semantic levels (e.g., domains such as math or code), in contrast to the low-level syntactic specialization observed in standard MoEs. Altogether, our results demonstrate a path toward modular, memory-efficient deployment of large, sparse models and open new opportunities for composable architectures.

Ryan Wang, Akshita Bhagia, Sewon Min

University of California, Berkeley · Allen Institute for AI

Language Model Pretraining Mixture-of-Experts Language Models Mixture of Experts Mixture-of-Experts Inference

May 7, 2026cs.LG

Federation of Experts: Communication Efficient Distributed Inference for Large Language Models

Mixture of experts has emerged as the primary mechanism for making Large Language Models (LLMs) computationally efficient. However, in distributed settings, communicating token embeddings between experts is a significant bottleneck. We present the novel Federation of Experts (FoE) architecture. FoE restructures the MoE block of a transformer layer into multiple MoE clusters. Each cluster is responsible for only one of the KV heads and expert parallelism is applied between those experts. Between clusters, a sum synchronizes the post-attention residuals, which then drives routing and dispatch for the next MoE block. In a single-node setting, FoE completely eliminates all-to-all communication as all experts within a group are contained on the same GPU. In multi-node settings, FoE confines all-to-all communication to the intra-node fabric, thus significantly reducing communication overhead. An implementation of FoE finds that on LongBench, FoE significantly improves inference throughput and latency in both single-node and multi-node settings, reducing the end-to-end forward-pass latency by up to 5.2x, TTFT by 3.62x, and TBT by 1.95x. It does so while achieving comparable generation quality to a mixture of experts model of the same size and training configuration.

Muhammad Shahir Abdurrahman, Chun Deng, Azalia Mirhoseini +1

Stanford University

Expert Parallelism Mixture-of-Experts Inference LLM Inference Acceleration Distributed Inference

May 7, 2026cs.DC

Relay Buffer Independent Communication over Pooled HBM for Efficient MoE Inference on Ascend

Mixture-of-Experts (MoE) inference requires large-scale token exchange across devices, making dispatch and combine major bottlenecks in both prefill and decode. Beyond network transfer, routing-driven layout transformation, temporary relay, and output restoration can add substantial overhead. Existing MoE communication paths are often buffer-centric, using explicit inter-process relay and reordering buffers around collective transfer. This report presents a relay-buffer-free communication design for MoE inference acceleration on Ascend systems. The design reorganizes dispatch and combine around direct placement into destination expert windows and direct reading from remote expert windows. Built on globally pooled high-bandwidth memory and symmetric-memory allocation, it removes most intermediate relay and reordering buffers while retaining only lightweight control state, including counts, offsets, and synchronization metadata. We instantiate the design as two schedules for the main phases of MoE inference: a prefill schedule with richer planning state for throughput-oriented execution, and a compact decode schedule for latency-sensitive execution. Experiments on Ascend-based MoE workloads show reduced dispatch and combine latency in both settings. At the serving level, the implementation improves time to first token (TTFT), preserves competitive time per output token (TPOT), and enlarges the feasible scheduling space under practical latency constraints. These results indicate that, on platforms with globally addressable device memory, reducing intermediate buffering and output restoration around expert execution is an effective direction for accelerating MoE inference.

Tianlun Hu, Tiancheng Hu, Shengsheng Litang +8

Huawei Technologies

Mixture-of-Experts Inference Inference Acceleration

May 7, 2026cs.LG

VisMMOE: Exploiting Visual-Expert Affinity for Efficient Visual-Language MoE Offloading

Large-scale vision-language mixture-of-experts (VL-MoE) models provide strong multimodal capability, but efficient deployment on memory-constrained platforms remains difficult. Existing MoE offloading systems are largely designed for text-centric workloads and become much less effective for visual-heavy inputs, where large numbers of visual tokens induce broader and less predictable expert accesses. We present VisMMoE, a VL-MoE offloading system built on a single systems insight: pruning redundant visual tokens can improve offloading not only by reducing computation, but also by reshaping expert demand. We refer to this effect as \textit{visual-expert affinity}: token pruning makes expert accesses more concentrated within layers and more stable across layers, producing a smaller and more predictable expert working set. Guided by this insight, VisMMoE combines affinity-aware token compression, lookahead expert prediction, and cache/pipeline orchestration to improve expert locality and prefetch effectiveness under tight memory budgets. We implement VisMMoE on multiple frameworks and evaluate it on representative VL-MoE models and benchmarks. VisMMoE improves end-to-end inference performance by up to 2.68x and 1.61x, respectively, over strong baselines for today's VL-MoE deployments while maintaining competitive accuracy.

Cheng Xu, Xiaofeng Hou, Jiacheng Liu +1

Shanghai Jiao Tong University Minhang Qu, Shanghai, China

Expert Offloading Vision-Language Models Efficient VLM Inference Mixture-of-Experts Inference Visual Token Compression

May 6, 2026cs.LG

Adaptive Inverted-Index Routing for Granular Mixtures-of-Experts

Mixture-of-experts (MoE) models enable scalable transformer architectures by activating only a subset of experts per token. Recent evidence suggests that performance improves with increasingly granular experts, i.e., many small experts instead of a few large ones. However, this regime substantially increases routing cost, which can dominate computation. We introduce adaptive inverted-index routing for MoE (AIR-MoE), an inverted-index-inspired routing architecture based on vector quantization (VQ). In a first stage, AIR-MoE performs coarse shortlisting by assigning tokens to VQ codewords to construct a candidate set of experts. In a second stage, fine scoring computes exact routing scores restricted to this shortlist. This two-stage procedure approximates true top-k routing while avoiding full expert scoring and, in contrast to prior work, imposing no structural constraints on expert parameters. AIR-MoE serves as a drop-in replacement for standard routers and requires no modifications to the model architecture or loss function. We further provide a lower bound on the mass recall achieved by AIR-MoE that yields insights into its inner workings. Empirically, we demonstrate that AIR-MoE achieves improved performance compared to existing routing approaches in granular MoE settings.

Klaus-Rudolf Kladny, Maximilian Mordig, Bernhard Schölkopf +1

Max Planck Institute for Intelligent Systems, Tübingen · Tübingen AI Center · ETH Zurich +1

Mixture-of-Experts Inference Expert Routing Mixture-of-Experts Models

May 3, 2026cs.LG

MoE-Prefill: Zero Redundancy Overheads in MoE Prefill Serving

Production LLM workloads increasingly serve discriminative tasks, such as classification, recommendation, and verification, whose answers are read from the logits of a single prefill pass with no autoregressive decoding. Serving these prefill-only workloads on mixture-of-experts (MoE) models is bottlenecked not by compute but by the distributed execution required to fit the model: existing parallel strategies (tensor, expert, and pipeline parallelism) trade memory pressure for redundant computation, communication, and synchronization, severely degrading MoE prefill serving efficiency. We observe that these overheads stem from coupling expert placement with synchronous activation routing -- a design inherited from the decoding era. The long, compute-bound forward passes of large-batch prefill open a per-layer window wide enough to stream expert weights in the background, replacing per-layer activation AllToAll with asynchronous weight AllGather fully overlapped with computation. We propose MoE-Prefill, a prefill-only serving system whose backend, AsyncEP (Asynchronous Expert Parallelism), gathers experts by weight rather than routing them by activation, and whose frontend co-enforces a physically-derived saturation threshold through prefix-aware routing and true-FLOPs load tracking. On Qwen3-235B-A22B across four hardware/precision configurations, MoE-Prefill delivers 1.35-1.37x throughput over the strongest distributed baseline on real-world workloads and up to 1.59x on long-context synthetic workloads, sustaining 29.8-36.2% per-GPU model FLOPs utilization.

Zhaoyuan Su, Olatunji Ruwase, Karthik Ganesan +5

University of Virginia · 2Snowflake AI Research · 3Harvard University

Expert Parallelism LLM Serving Mixture-of-Experts Inference Distributed Inference

May 1, 2026cs.LG

Affinity Is Not Enough: Recovering the Free Energy Principle in Mixture-of-Experts

Sparse MoE routing fails at domain transitions, where the current token belongs to one distribution and the next to another. In a controlled experiment (4 experts, 5 seeds), standard affinity routing assigns only 0.006 +/- 0.001 probability to the correct expert at the transition. Three lightweight gate modifications raise this to 0.748 +/- 0.002 (124x), cutting experts needed for 99% coverage from infeasible to a small constant: temporal memory (beta), a per-expert LIF membrane potential accumulating routing context across tokens; precision-weighted gating (Pi), a per-expert inverse variance of recent prediction error, yielding 31x contrast between reliable and unreliable experts; and anticipatory routing, a next-state predictor conditioned on the beta-accumulated hidden state. The mechanisms draw from Friston's Free Energy Principle and use LIF dynamics from spiking neural networks. An ablation across all 2^3 subsets reveals a super-additive beta x Ant interaction: anticipation alone gives nothing (+0.000 +/- 0.001); beta alone gives modest gain (+0.295 +/- 0.013); combined they close 75% of the oracle gap (+0.741 +/- 0.002, exceeding the sum by +0.446 +/- 0.014). This is structural: a stateless predictor cannot detect approaching transitions because pre-transition tokens are distributionally identical to within-domain tokens. In a character-level MoE LM (5 seeds), beta-routing reduces transition-step BPC from 6.56 +/- 0.01 (Standard) to 4.01 +/- 0.15 (beta-MoE); the beta + Ant gate places 0.86 +/- 0.02 probability on the correct domain expert before that domain appears in input, vs 0.42 +/- 0.12 for Standard MoE. Reference implementations (~200 lines each): https://github.com/russellwmy/affinity-is-not-enough

Man Yung Wong

Russell · Independent Researcher

Sparse Mixture-of-Experts Mixture-of-Experts Inference Free Energy Principle Expert Routing