Mixture-of-Experts Models

Follow this topic Ask about this topic

Momentum

41 papers in the last four weeks, up 116% on the four weeks before. 0.4% of all new papers.

Jul 13Week of Sep 28

Latest papers 297

Jul 28, 2026cs.LG

Spend Experts Where You Are Unsure: Confidence-Adaptive Routing for Mixture-of-Experts LoRA

Mixture-of-Experts (MoE) variants of Low-Rank Adaptation (LoRA) route every token to a fixed number of experts

k

. Tokens differ in how uncertain the model is about them, so a single k over-spends on easy tokens and under-serves hard ones. We observe that the router's output distribution is already a per-token uncertainty signal: peaked mass indicates confidence, while a flat distribution indicates ambiguity. We introduce CARE (Confidence-Adaptive Routing of Experts), which admits experts in a nucleus fashion. Experts are activated in decreasing router weight until their cumulative mass reaches a threshold, with a small extension when the admitted experts disagree. A budget thermostat calibrates the threshold so that the average number of active experts matches any target. CARE is a drop-in, single-forward-pass rule with no extra parameters. Across eight commonsense benchmarks on LLaMA-3.1-8B and Qwen2.5-7B, as well as math, code, and knowledge tasks, CARE improves over fixed top-k MoE-LoRA at matched compute and matches the fixed-k=4 baseline while activating fewer experts. The same confidence and disagreement signals also improve out-of-distribution detection over MSP, entropy, and multi-pass proxies. We support the design with nucleus fidelity, budget optimality, and an epistemic reading of disagreement, and we release code.

Tom Saliencro, Rohan Desai, Priya Nair +2

University of California, Irvine · University of Washington

Parameter-Free Mixture-of-Experts Routing LLM Inference Acceleration Expert Routing LLM Uncertainty Estimation Mixture-of-Experts Models

Jul 28, 2026cs.CV

Balanced Soft mixture-of-expert model for Glaucoma Detection

Glaucoma is a group of eye diseases that damage the optic nerve, often caused by elevated intraocular pressure. It is a leading cause of irreversible vision loss and is typically developed slowly and painlessly, making it difficult to notice until significant damage has occurred. Therefore, early detection is crucial to prevent or slow the progression of vision loss. In recent years, deep learning based uni-modal models have improved the accuracy and efficiency of glaucoma detection, empowering doctors with tools for earlier diagnosis, better monitoring, and timely treatment. Building on this, multi-modal models have emerged, leveraging the strengths of different imaging modalities to learn richer and more robust representations, further enhancing glaucoma detection accuracy. However, multi-modal learning faces challenges such as imbalanced and under-optimized uni-modal representations due to joint learning objectives. To address this, we propose a balanced soft mixture-experts model with three experts and load balancing loss. The performance is measured by AUC, our proposed method surpasses the performance of all uni-modal baselines, conventional multi-modal models, and current stateof- the-art balanced multi-modal models. The proposed model can be generalized to other disease detections such as diabetic retinopathy.

Sai Venkatesh Chilukoti, Krishna Rauniyar, Min Shi +1

School of Computing and Informatics, University of Louisiana at Lafayette, 301 East Lewis Street, Lafayette, 70504, LA, USA.

Multimodal Classification Mixture-of-Experts Models Multimodal Fusion

Jul 27, 2026cs.CV

MMOE: Modernizing Diffusion Transformers with Efficient Expert Design

Modern large language models scale successfully by pairing capacity growth with efficiency, keeping per-token and deployment costs under control as capacity grows. AIGC Foundation Models (AFMs), especially diffusion-transformer backbones, have begun to adopt sparse experts, but recent efforts mostly enlarge total parameter counts and sparsity ratios without importing the efficiency mechanisms that made LLM scaling practical, so generation quality is seldom balanced against training and deployment cost. This raises a natural question: can the architectural principles behind efficient LLM scaling be adapted to AFMs in a more balanced way? We introduce ModernMOE (MMOE), a modernization of SiT-style diffusion transformers that systematically adapts routed experts, shared and lightweight experts, gate-residual routing, and attention-residual information reuse to AIGC generation. Rather than treating MoE as a single plug-in replacement, MMOE studies how different modern expert components affect convergence, efficiency, and generation quality when composed inside a diffusion transformer. Every experiment in this paper is trained on a single eight-GPU H100 node with batch size 256 for 400k steps, an accessible single-machine budget. Under matched training and sampling protocols and at this budget, MMOE reaches lower FID at every recorded checkpoint, that is, it converges faster per training step, than dense and intermediate sparse-expert baselines, and among the sparse variants it attains the best quality-cost balance. Routing analysis further shows stable expert specialization across depth, substantial use of lightweight routes, and modest step-to-step routing changes during denoising. These results suggest that AFMs can follow the balanced scaling path of LLMs by importing proven efficiency designs, rather than by simply increasing total parameters and sparsity ratios.

Yanhao Jia, Jiepeng Wang, Haibin Huang +3

College of Computing and Data Science, Nanyang Technological University, Singapore, 639798 · Institute of Artificial Intelligence of China Telecom (TeleAI), Shanghai, China, 200232

Diffusion Transformer Diffusion Model Acceleration Mixture-of-Experts Models Diffusion Models

Jul 25, 2026cs.LG

Hierarchical Copula-Gumbel-Top-\texorpdfstring{ $K$ }{K} Routing: Two-Sided Dependence Control for Frozen Mixture-of-Experts at Fixed Per-Token Routing Laws

A stochastic Gumbel-Top-

K

router defines, for every token of a mixture-of-experts (MoE) model, a \emph{routing law}: a distribution over ordered expert lists and mixture weights. We ask which \emph{joint} distributions over the routing choices of different tokens are reachable while every individual token's complete routing law is held exactly fixed. We give a two-sided construction, \emph{Hierarchical Copula-Gumbel-Top-

K

} (\CGA{}). Within a group of related tokens, an exchangeable Gaussian copula positively correlates the Gumbel perturbations at each expert coordinate, which can increase within-group expert-set coherence. Across disjoint pairs of groups, a tunable antithetic construction introduces a selectable amount of negative dependence. We prove that both operations leave each token's ordered Top-

K

sample, mixture weights, and inclusion probabilities identical in distribution to independent routing \emph{at a routing layer conditioned on its pre-routing logits}; conditional expected expert traffic is preserved as a consequence. We characterize the resulting trade-off: positive within-group coupling can only inflate the variance of realized expert loads relative to independent routing, while nonnegative cross-group opposition can only reduce it relative to flat coupling at the same within-group strength. Coherence and load dispersion are thus controlled by two complementary dependence dials on the invariance constraint surface. Because the base model is untouched, the dials can be driven by a small controller over frozen features, trainable with a score-function estimator: the frozen network is evaluated only in the forward direction, and gradients are confined to the controller. An initial small-scale pilot validates the mechanism and the training route, but does not establish task-level fine-tuning gains.

Richard Yi Da Xu

Hong Kong Baptist University and TadReamk Limited

Copula Models Expert Load Balancing Parameter-Free Mixture-of-Experts Routing Expert Routing Mixture-of-Experts Models

Jul 22, 2026cs.RO

Emergent Compositional Skills in Mixture-of-Experts VLAs

We consider the problem of learning compositional robot policies end-to-end from expert demonstrations, without any pre-specified notion of task decomposition or hierarchy. We ask whether a VLA trained with a simplified Mixture-of-Experts (MoE) action head can emergently learn to decompose tasks into reusable, interpretable primitives. We find that learned experts are heavily reused across tasks and consistently correspond to qualitatively distinct low-level behaviors, suggesting that the router implicitly learns to perform high-level sequencing while experts serve as compositional primitives. Our MoE matches the task performance of a monolithic baseline while demonstrating meaningful expert specialization, a step toward modular, interpretable robot policies that emerge from data alone.

Shlok Shah, Rhiaan Jhaveri, Tharun Kumar Tiruppali Kalidoss +3

Department of Computer Science, Princeton University, Princeton, NJ, USA · Department of Electrical and Computer Engineering, Princeton University, Princeton, NJ, USA

Robot Policy Learning Robot Skill Learning Robot Imitation Learning Agent Skill Learning Interpretability in RL Mixture-of-Experts Models+1

Jul 22, 2026cs.CV

MoAKE: Toward Unified All-in-One Action Quality Assessment via Mixture of Action Knowledge Experts

Action Quality Assessment (AQA) aims to objectively evaluate performance quality from action videos. Most existing methods follow a ``one-by-one'' paradigm, training a separate model for each action type. This setting limits real-world deployment, as it requires prior action-type knowledge to select the corresponding model and suffers from poor generalization across diverse actions. To address these limitations, we study the challenging task of all-in-one AQA, which aims to assess heterogeneous actions within a single unified model. We propose a novel Mixture of Action Knowledge Experts (MoAKE) framework, designed to mitigate negative knowledge transfer caused by large semantic discrepancies among actions. MoAKE learns complementary experts that capture diverse action patterns within a shared semantic space and dynamically aggregates their knowledge to adapt the assessment to the input action. Each expert is tailored with segment-aware prototypes to handle varying temporal lengths, together with an Adaptive Intra- and Inter-Segment Relationship Modeling (AIISRM) module to model multi-granularity temporal dynamics. Furthermore, we establish comprehensive benchmarks for all-in-one as well as zero/few-shot AQA. Extensive experiments on three long-term datasets demonstrate that MoAKE significantly outperforms existing methods in the all-in-one setting, while also achieving consistent generalization on three short-term datasets under zero/few-shot evaluation. Code is available at https://github.com/XuHuangbiao/MoAKE.

Huangbiao Xu, Huanqi Wu, Xiao Ke +3

Fujian Provincial Key Laboratory of Networking Computing and Intelligent Information Processing, College of Computer and Data Science, Fuzhou University, Fuzhou 350108, China · Engineering Research Center of Big Data Intelligence, Ministry of Education, Fuzhou 350108, China · School of Intelligence Science and Technology, University of Science and Technology Beijing, Beijing 100083, China

Mixture-of-Experts Models Action Quality Assessment

Jul 21, 2026cs.CV

Continual Video-MLLM Adaptation over Evolving Domains

Video multimodal large language models have shown strong capability in video understanding, yet their adaptation to sequentially evolving domains remains underexplored. In real-world deployments, video data often arrives continuously from heterogeneous domains, requiring the model to acquire new domain-specific knowledge without overwriting previously learned capabilities. Existing continual learning methods typically rely on shared adaptation spaces, which can induce severe cross-domain interference and catastrophic forgetting. We propose Distribution-Aware Expert Routing, a parameter-efficient framework for continual Video-MLLM adaptation over evolving domains. DAER maintains domain-isolated lightweight experts while keeping the pretrained Video-MLLM backbone frozen, thereby decoupling domain-specific adaptation from the general multimodal knowledge of the pretrained model. To enable fine-grained specialization, we introduce an intra-domain distribution-aware routing mechanism that matches each input to expert-level prototype reservoirs using MMD. To address the absence of task identities at inference time, we further propose an inter-domain routing mechanism that performs prototype matching in a discriminative subspace for robust domain identification. In addition, we introduce adaptive domain merging to improve parameter scalability and adopt a two-stage optimization strategy to stabilize expert specialization during continual learning. We evaluate DAER by curating a domain-incremental benchmark built from ten VidQA datasets covering diverse visual environments and reasoning demands. Experiments on two strong Video-MLLM backbones show that DAER consistently outperforms prior methods.

Rui Cheng, Meixing Shi, Yuxiang Cai +3

School of Software Technology, Zhejiang University, Ningbo, China · Zhejiang Key Laboratory of Digital-Intelligence Service Technology, China · Hong Kong Polytechnic University, Hong Kong SAR +1

Video QA Domain-Incremental Learning Mixture-of-Experts Models Continual Learning for VLMs

Jul 20, 2026cs.LG

Adaptive Multi-Expert Graph Transformer for Interpretable EEG-Based Diagnostics

Electroencephalographic (EEG) abnormalities arise from dynamic changes in neural synchrony across spatial and temporal scales, yet many computational approaches reduce these dynamics to static features. We present a Spatial Multi-Expert Graph Transformer that models each EEG recording as a sequence of dynamic functional connectivity graphs. Time-resolved connectivity is estimated using the weighted Phase Lag Index (wPLI), and hierarchical graph encoding aggregates information from electrode to regional and global levels. A multi-expert transformer architecture enables subtype-aware reasoning, with a gating mechanism adaptively fusing expert outputs for global abnormality prediction. Experiments on the TUAB dataset show competitive abnormal EEG detection performance and demonstrate the potential of dynamic graph modeling with adaptive expert fusion for interpretable, subtype-aware spatial--temporal analysis.

Maryam Rahimimovassagh, Md Elias Hossain, Ivan Garibay +1

Department of Industrial Engineering and Management Systems University of Central Florida, Orlando, FL, USA

Electroencephalography Dynamic Graph Learning Mixture-of-Experts Models Graph Transformer Anomaly Detection Brain Network Analysis

Jul 18, 2026cs.LG

HyBDM: Multi-Scale Hybrid Experts for Time Series Forecasting with Bidirectional Dependency Modeling

Time series forecasting (TSF) is vital to many applications, yet existing models often struggle to capture the heterogeneous long-range global patterns and short-range local variations in multivariate time series. While some approaches partially model these dependencies, they often do not jointly exploit temporal and feature-wise information. To address this challenge, we propose HyBDM, a multi-scale hybrid model that decomposes temporal dynamics into global patterns and local variations, which are modeled by two specialized experts. The Global Patterns Expert employs an enhanced BiConv-Mamba module that integrates bidirectional convolutions, an M-SSM layer, a forgetting mechanism, and a GDD-MLP module for cross-channel modeling. The Local Variations Expert uses a Local Window Transformer (LWT) to perform efficient locality-aware attention with reduced computational complexity. In addition, a Multi-Scale Patcher and a Long-Short Router enable multi-resolution representations and adaptive fusion of the two experts. Experiments on six benchmark datasets show that HyBDM outperforms state-of-the-art methods in both forecasting accuracy and computational efficiency, demonstrating its effectiveness in bridging global-local dependencies for multivariate TSF.

Wenqiang Ma, Chen Cheng, Xue Cheng +1

Soul of Harlequin Lab

Multivariate Time Series Forecasting Multi-Scale Feature Fusion Mixture-of-Experts Models Time Series Transformers Multiscale Time-Series Modeling

Jul 18, 2026cs.CV

Can Experts Adapt Without Training? On Test-Time Modality Generalization in MVLMs

Medical vision-language models (MVLMs) promise broad zero-shot generalization, yet their reliability collapses when confronted with unseen modalities and domains, precisely where clinical robustness matters most. To address this gap, we revisit test-time modality generalization from the perspective of Mixture-of-Experts (MoE) and ask: can experts route-and-adapt without any optimization during inference? We identify a fundamental specialization-generalization dilemma at test time, where blindly aggregating modality experts dilutes modality-specific knowledge, while selecting one highly confident expert risks mismatch under shift. To address this, we propose MoBE: a fully optimization-free framework that performs dynamic expert selection and adaptation at test time. MoBE combines entropy-guided dynamic routing in MoE settings with expert-wise Bayesian adaptation, enabling experts to update their confidence and adapt online without gradient updates. Without parametric updates, MoBE augments a static MVLM with test-time routing and online statistics, achieving average accuracy gains of +4.72, +7.17, and +4.3 over state-of-the-art TTA methods across seen, unseen, and heterogeneous medical benchmarks, highlighting the effectiveness of training-free expert adaptation for robust modality generalization.

Raza Imam, Darakshan Rashid, Yutong Xie +3

Mohamed bin Zayed University of Artificial Intelligence, Abu Dhabi, UAE · Indian Institute of Technology Delhi, India · Khalifa University, Abu Dhabi, UAE

VLM Robustness VLM Adaptation Test-Time Adaptation of VLMs Medical VLMs Mixture-of-Experts Models

Jul 15, 2026cs.CV

MixCompress: Mixture of Experts for Variable Rate Learned Image Compression

Learned image compression (LIC) is bottlenecked by the need to store independent models for each rate-distortion operating point. Existing variable bit-rate (VBR) methods aim to reduce this overhead via dense parameter modulation, but forcing a shared backbone to approximate divergent mappings causes severe feature entanglement. Specifically, low-rate smoothing gradients inherently conflict with the preservation of high-frequency textural details, leading to sub-optimal performance. To resolve this, we propose MixCompress, a unified VBR framework based on sparse structural specialization. While sparsely gated Mixture-of-Experts (MoE) routing successfully mitigates gradient conflict, it operates on a fixed computational budget. To address the increased representational demands of higher bit-rates we introduce a Mixture-of-Depths (MoD) extension to dynamically scale model capacity. Combined with Conditional Auxiliary Transforms (CAT) for dynamic sub-band energy modulation, our hierarchical framework effectively dynamically scales capacity. Extensive evaluations demonstrate that MixCompress not only matches individually optimized single-rate baselines but can even surpass them, establishing a new Pareto frontier for computationally efficient image coding.

Calvin-Khang Ta, Praneet Singh, Tong Shao +1

Dolby Laboratories, USA

Learned Image Compression Rate-Distortion Optimization Mixture-of-Experts Models

Jul 15, 2026cs.CV

Multi-Expert Routing for Multi-Domain Low-Resource OCR: A Manchu Case Study

Historical Manchu OCR must accommodate various visually distinct writing styles, including regular script, running script, and the semi-cursive chancery hand used in palace memorials, despite limited labeled data. We study a multi-expert system that reuses checkpoints from an iterative fine-tuning process as domain specialists and uses a lightweight page-level image classifier to dispatch pages by visual style. When the checkpoint pool lacks a suitable specialist, we train an additional expert for that domain. On three frozen test sets, the routed system matches the selected specialist for each style at two-decimal precision: 0.30 percent CER on regular script, 1.57 percent on memorials, and 4.83 percent on running script. The router achieves 99.3 percent page-level domain accuracy and matches the domain-label oracle at the same precision. Two of the three selected specialists were not trained specifically for their final domain; only the running-script expert was trained with that domain as its target. We report the evaluation protocol, router design, and per-page predictions to make the comparison reproducible.

Zhan Chen, Jiqiao Ma, Chih-wen Kuo

Institute of Advanced Studies, Beijing Normal University, Zhuhai, China · Faculty of Humanities and Social Sciences, Beijing Normal University–Hong Kong Baptist University United International College (UIC), Zhuhai, China · Independent Scholar +1

Mixture-of-Experts Models Optical Character Recognition

Jul 14, 2026cs.CL

Less Experts, Faster Decoding: Cost-Aware Speculative Decoding for Mixture-of-Experts

Sparse Mixture-of-Experts (MoE) models have become an important approach for scaling Large Language Models (LLMs), but their inference efficiency depends strongly on expert activation patterns. Speculative decoding (SD) accelerates autoregressive generation by verifying multiple draft tokens in parallel, yet existing draft selection strategies primarily optimize acceptance likelihood. In large-scale MoE models, however, selecting draft tokens also determines the union of experts activated during verification. We observe that confidence-driven SD can introduce \textit{expert scattering}: high-probability draft tokens may route to disjoint experts, increasing expert-weight memory traffic and reducing the speedup from speculation. Motivated by this observation, we revisit draft-tree selection under the non-uniform memory-cost structure of MoE inference. We propose \textsc{EcoSpec}, a cost-aware speculative decoding framework that incorporates predicted marginal expert activation cost into draft selection. With a lightweight expert predictor and a dynamic expert buffer, \textsc{EcoSpec} favors draft paths that preserve high acceptance likelihood while reusing experts already covered by the current verification set, without modifying the target-model verification rule. We evaluate \textsc{EcoSpec} on three large-scale MoE models, including DeepSeek-V3.1 (671B), Qwen3-235B-A22B, and GPT-OSS-120B, across reasoning, coding, question-answering, and dialogue benchmarks. \textsc{EcoSpec} consistently reduces active expert footprints and improves end-to-end decoding speed, achieving up to

1.62\times

speedup. These results show that accounting for expert activation cost is important for efficient speculative decoding in large-scale MoE models.

Jincheng Xie, Runheng Liu, Heyan Huang +4

Department of Mathematical Sciences, Tsinghua University, Beijing, China · School of Computer Science and Technology, Beijing Institute of Technology, Beijing, China · JDT AI Infra +2

Mixture-of-Experts Inference Speculative Decoding Mixture-of-Experts Models

Jul 14, 2026cs.LG

Understanding Structured Health Data through Interaction-Aware Mixture-of-Experts

We study interaction-aware mixture-of-experts for post-stroke rigidity prediction using multi-level views of structured health records. Despite minimal performance gains, routing attribution reveals systematic importance differences across views, underscoring view construction as key to interpretability.

Ji Hwan Park, Ying Ding, Tianjin Guo

Department of Electrical and Computer Engineering, The University of Texas at Austin, Austin, Texas, USA · School of Information, The University of Texas at Austin, Austin, Texas, USA · Kelley School of Business, Indiana University, Bloomington, Indiana, USA

Multi-View Learning Clinical Outcome Prediction Neural Network Interpretability Mixture-of-Experts Models

Jul 13, 2026cs.LG

From Global to Factor-Wise Expert Composition in Discrete Diffusion Models

Discrete diffusion models offer a powerful framework for solving complex reasoning tasks, particularly through compositional generation, which combines multiple pre-trained experts to generalize beyond their individual training data. Recent theoretical corrections introduce time-dependent mixing weights to better align composed diffusion dynamics with the intended target. However, these methods are fundamentally limited by working on a per-sample basis, treating each generated state monolithically and ignoring the potential spatial or functional specializations of different experts. In this work, we address this limitation by proposing FactorDiff - a factor-wise composition framework for diffusion models. We posit that samples can be further decomposed into smaller factors, and propose a sampling process that dynamically routes each factor to the most relevant expert. We instantiate this framework with spatial/pixel-level compositions and validate it on the ARC-AGI benchmark, demonstrating that simple factor-specific routing consistently outperforms complex global scalar weighting schemes on tasks that require logical consistency and spatial disentanglement.

Haozhe Huang, Yudong Xu, Abhijoy Mandal +1

Department of Computer Science, University of Toronto · Vector Institute for Artificial Intelligence · Department of Mechanical & Industrial Engineering, University of Toronto +1

Expert Routing Mixture-of-Experts Models Diffusion Models

Jul 13, 2026cs.AI

HCRMap: Pressure-Aware Hot-Expert Residency Mapping for 3.5D MoE Chiplet Inference

Mixture-of-Experts (MoE) large language models (LLM) activate only a small number of experts during inference, but token routing introduces persistent expert hotness skew: a small set of hot experts continuously receives most tokens, while the remaining experts are lightly loaded. On 3.5D multi-chiplet systems, this skew not only causes compute imbalance but also amplifies pressure on communication, memory bandwidth, I/O, and execution queues. Therefore, the core problem is not simply to reduce token movement, but to dynamically place and reuse hot expert replicas across different memory tiers. This paper proposes HCRMap, a hot expert residency mapping framework for pressure-aware expert replica management in 3.5D MoE inference. Based on expert hotness, weight loading cost, migration overhead, and runtime resource pressure, HCRMap dynamically determines which experts should be promoted, retained, demoted, or evicted. It then maps routed token groups to suitable resident replicas, thereby jointly mitigating communication, memory, and queue bottlenecks. Experimental results show that HCRMap reduces end-to-end latency by 43.6% and 43.0% over Hydra in the prefill and decode stages, respectively; by 34.5% and 33.1% over MoEntwine; and by 46.7% and 46.0% over PIMoE.

Yongqin Zhang

Nanjing Vocational College of Information Technology, 99 Wenlan Road, Xianlin College Community, Nanjing, Jiangsu, 210023, P.R.C.

AI Accelerator Inference Mixture-of-Experts Inference Mixture-of-Experts Models

Jul 10, 2026cs.LG

Mach-Mind-4-Flash Technical Report

We present Mach-Mind-4-Flash, a 35B-parameter Mixture-of-Experts (MoE) agentic model with 3B activated parameters. Through post-training optimization alone without scaling pre-training compute, the model achieves performance on par with or surpassing that of 100B-parameter-class models. By introducing scalable agentic interaction environments for large-scale reinforcement learning, the model attains significant performance gains on real-world application tasks. Our pipeline comprises three stages: (1) a unified RL/OPD training infrastructure with dynamic multi-teacher scheduling and operator-level acceleration, delivering 17% end-to-end training speedup; (2) multiple domain-specific RL experts trained in parallel across Reasoning, General, and Agent tracks, then fused into a single generalist via Multi-Teacher On-Policy Distillation (MOPD) -- a routed reverse-KL objective that eliminates the see-saw degradation of mixed-reward RL; (3) Hybrid Median-length Policy Optimization (HMPO), a single-stage token-efficiency method that compresses reasoning chains by 19--46% with

\le

0.7 percentage-point accuracy loss. Mach-Mind-4-Flash scores 92.70 on AIME'26, 82.82 on IFBench, 80.74 on Behavioral-SafetyBench, 75.80 on BFCL-v4, 72.31 on BrowseComp-zh, and 84.20 on ClawBench -- leading or matching models with 10--30

\times

its activated size at a fraction of the inference cost.

Foundation Model Team

Foundation Model, Li Auto Inc.

Agentic RL Token Efficiency LLM Agent Training On-Policy Distillation RL Post-Training Mixture-of-Experts Models

Jul 9, 2026cs.CL

It Takes a MAESTRO To Prune Bad Experts

Sparsely-activated Mixture-of-Experts (MoE) language models achieve remarkable inference efficiency by activating only a small fraction of parameters per token, yet their full expert banks reside in memory at all times, creating a prohibitive deployment bottleneck. Existing structured pruning methods, largely designed for dense transformers, assess expert importance using locally derived heuristics that are blind to the interdependent nature of MoE routing. We introduce MAESTRO (Markov-chain Approximated Expert Sparsification via Transition-based ROuting), a structured pruning framework designed for MoE architectures that models autoregressive expert activation trajectories as Ergodic Markov chains whose stationary distributions encode cross-layer dependencies, yielding a globally aware importance heuristic. Evaluated across five diverse domains including Safety, Bias, and Ethics, MAESTRO outperforms state-of-the-art baselines by up to 10.61% in average performance retention under a strict 50% compression regime, while exhibiting substantially lower cross-task variance, indicating that global, routing-congruent pruning produces models that generalize more consistently across heterogeneous tasks.

Palaash Goel, Ayush Maheshwari, Tanmoy Chakraborty

Indian Institute of Technology Delhi, India · NVIDIA, India

Mixture-of-Experts Pruning LLM Pruning Mixture-of-Experts Language Models LLM Compression Activation Sparsity Structured Sparsity+1

Jul 9, 2026cs.AI

SHAP-Weighted Cross-Modal Expert Fusion for Emotion and Sentiment Recognition: Evidence and Limits

Multimodal emotion and sentiment recognition is commonly addressed by early fusion, which concatenates modalities before classification, or late fusion, which combines independently trained unimodal predictors. Early fusion can be accurate but monolithic, while late fusion is modular but may lose cross-modal interactions. This paper revisits XAI-guided adaptive fusion (\xgaf), a tree-based mixture of unimodal and cross-modal experts whose sample-level weights are derived from TreeSHAP attribution magnitudes. We focus on the effect of SHAP attribution reduction when experts have unequal feature dimensionalities. In this setting, mean-abs and median-abs reductions can suppress high-dimensional cross-modal experts, whereas sum-abs reduction preserves total attribution mass. On MELD 7-class emotion recognition, sum-abs \xgaf{} nearly matches early fusion across three face-sequence aggregators; the Transformer variant reaches 0.5983 \wf{}, compared with 0.6018 for early fusion and 0.4598 for probability-average late fusion. McNemar testing shows no significant difference between sum-abs \xgaf{} and early fusion on MELD (

p=1.000

), while \xgaf{} remains significantly better than late fusion (

p<0.0001

). On CMU-MOSEI 3-class sentiment recognition, sum-abs \xgaf{} reaches 0.6519 \wf{}, slightly exceeding early fusion (0.6485) and late fusion (0.5696). Ablation studies show that the main gain comes from adding cross-modal experts, especially the trimodal expert, rather than from complex per-sample routing. Diagnostics further show that mean-abs and median-abs weights are nearly uniform, while sum-abs weights concentrate on the trimodal expert. Thus, the main contribution is a transparent empirical analysis of how SHAP reduction, expert dimensionality, and cross-modal expert design affect modular multimodal fusion.

Adis Alihodzic, Selma Skopljakovic Hubljar

Department of Mathematical and Computer Sciences, Faculty of Science, University of Sarajevo, Sarajevo, Bosnia and Herzegovina.

SHAP Feature Attribution Multimodal Emotion Recognition Multimodal Sentiment Analysis Mixture-of-Experts Models Multimodal Fusion Multimodal Understanding

Jul 9, 2026cs.CV

On the Design of Mixture-of-Experts for Dynamic Gaussian Splatting

Dynamic scene reconstruction remains challenging due to the heterogeneous and spatially varying nature of real-world motion. Although recent 3D Gaussian Splatting methods have introduced diverse deformation formulations for dynamic novel view synthesis, each method typically relies on a single deformation model within its representation, which limits robustness across diverse dynamic scenarios. In this work, we study a fundamental problem-multi-deformation modeling for dynamic 3D Gaussian representations-under two distinct integration constraints that differ in when and how multiple deformation experts interact during training. From a Mixture-of-Experts (MoE) perspective, we view multi-deformation modeling as the problem of combining multiple specialized deformation models within a unified 3D representation. We first introduce Mixture of Deformation Experts (MoDE), which integrates multiple deformation experts directly into the deformable Gaussian Splatting pipeline through joint optimization. In MoDE, experts operate on a shared canonical Gaussian representation, enabling multi-deformation modeling without introducing additional training stages or modifying the original optimization schedule. In contrast, we further present Mixture of Experts for Dynamic Gaussian Splatting (MoE-GS) under a different integration constraint, where deformation experts are optimized independently and combined through a separate routing stage. As a result, expert interaction occurs over non-canonical Gaussian representations after individual optimization. Together, these two approaches provide alternative strategies for multi-deformation modeling, clarifying how integration constraints shape the design and behavior of deformation experts in dynamic 3D Gaussian representations. Our code is available at: https://github.com/cvsp-lab/MoE-GS-studio.

In-Hwan Jin, Hyeongju Mun, Joonsoo Kim +2

Department of Electrical and Electronics Engineering, Pusan National University, Busan, Republic of Korea · Immersive Media Research Section, Electronics and Telecommunications Research Institute (ETRI), Daejeon, Republic of Korea · Department of Electrical and Electronics Engineering and JYS AI Nexus Institute, Pusan National University, Busan, Republic of Korea

3D Gaussian Splatting Dynamic Scene Reconstruction Dynamic Novel View Synthesis Mixture-of-Experts Models

Jul 9, 2026cs.CV

Mixture of Enhanced-View Experts for Multi-Query Vehicle ReID and A Large-Scale Benchmark

Multi-query vehicle ReID aims to leverage complementary information from diverse views for robust feature learning. However, current methods suffer from simplistic feature fusion and thus easily ignores some important view information and cross-view relationships. To handle these problems, this work presents a novel approach called Mixture of Enhanced-View Experts (EV-MoE), which enhances the feature representation of each view and efficiently integrate the view-specific enhanced features by MoE, for robust multi-query ReID. In particular, we design a mixture of enhanced-view experts module, which consists of two parts including view-specific feature enhancement sub-Module (VFEM) and dynamic multi-view fusion sub-Module (DMFM). Moreover, we further introduce Multi-view Alignment Loss (MAL), which aligns features through bidirectional crossview contrastive learning and reconstruction constraints, addressing the challenges of consistency between multi-query features and single-image features. In addition, to evaluate multi-query ReID in real-world environments, we collect LCRI-1K, a largescale vehicle ReID dataset with 1,090 identities, 107,805 images, across 23,637 cameras, where each vehicle appears in an average of 67.5 cameras, providing a comprehensive benchmark to test the robustness in complex environments. Extensive experiments demonstrate the robustness of CAFNet in addressing the multiquery vehicle ReID problem. The code is available at https: //github.com/xiaozhen28/CAFNet.

Aihua Zheng, Jie Zhen, Chenglong Li +2

Information Materials and Intelligent Sensing Laboratory of Anhui Province, Anhui Provincial Key Laboratory of Multimodal Cognitive Computation, School of Artificial Intelligence, Anhui University. Hefei, 230601, China · Anhui Provincial Key Laboratory of Multimodal Cognitive Computation, School of Artificial Intelligence, Anhui University. Hefei, 230601, China · School of Artificial Intelligence, Anhui University of Science and Technology, Hefei 232001, China +1

Multi-View Consistency Multi-View Learning Mixture-of-Experts Models

Jul 8, 2026cs.CV

Scaling Mixture-of-Experts Video Pretraining for Embodied Intelligence

Despite the recent promise in robot control, video generative models suffer from a domain mismatch due to their primary focus on content creation. For example, their design inherently prioritizes visual fidelity and creativity over computational efficiency and physical realism. In this work, we present LingBot-Video, a DiT-based video pretraining paradigm specifically tailored for embodied intelligence. From the architecture perspective, we adopt the Mixture-of-Experts (MoE), instead of dense, framework to achieve a better trade-off between modeling capacity and inference efficiency, and manage to scale it up from scratch. From the data perspective, we construct a data profiling engine that augments standard internet videos with extensive robot-oriented footage, encompassing manipulation, navigation, and egocentric perspectives, to equip the base model with an intrinsic understanding of actions and world dynamics. From the training perspective, we develop a multi-dimensional reward system to enforce the alignment regarding physical rationality and task completion, going beyond standard criteria such as aesthetics, prompt-following, and motion consistency. Comprehensive evaluations validate its performance and efficiency as a video foundation model. We contribute LingBot-Video as the inaugural large-scale, open-source MoE video foundation model to the community, in a pioneering effort to bridge digital creativity and physical actuation.

Shuailei Ma, Jiaqi Liao, Xinyang Wang +24

Video Diffusion Models Self-Supervised Pre-Training Video Generation Mixture-of-Experts Models

Jul 8, 2026cs.LG

Image classification via a quantum-inspired strategy involving a mixture of experts

Pattern recognition problems arise in a variety of physical image processing situations, and convolutional neural networks are a popular scheme for the required feature extraction and classification tasks. The classical networks use diffusion-based smearing and block-wise pooling to downsample the image data and capture important structural features. In this work, we propose and demonstrate a more efficient quantum-inspired strategy involving a mixture of experts. It is a hybrid classical-quantum framework. The quantum part consists of amplitude encoding of the images, convolution using local unitary operations, multiple experts processing the same image with different parameters, and feature extraction using quantum stabiliser codes. The classical part then jointly processes the features extracted by different experts using a standard fully connected neural network for image class prediction. Using MNIST and Fashion-MNIST datasets as benchmarks, we demonstrate that the joint expert analysis outperforms the individual expert one, as well as reduces the failure rate of image class prediction by around a factor of two. The overhead of our quantum-inspired strategy is only moderate on GPU workstations, which makes our proposal a practical alternative to existing classical schemes. We also point out how the quantum part of our framework can be executed on a quantum processor.

Kumari Jyoti, Rohith Babu, Apoorva D. Patel

Centre for High Energy Physics, Indian Institute of Science, Bangalore 560012

Hybrid Quantum-Classical ML Quantum-Inspired ML Mixture-of-Experts Models Image Classification

Jul 7, 2026cs.CV

ProMoE-FL: Prototype-conditioned Mixture of Experts for Multimodal Federated Learning with Missing Modalities

In this paper, we address the problem of multimodal federated learning with missing modality. Existing methods utilize an additional public dataset or perform naive feature synthesis that is based solely on the available modality. To address these limitations, we propose ProMoE-FL, a Prototype-conditioned Mixture-of-Experts framework for robust missing-modality feature synthesis in multimodal federated learning. ProMoE-FL builds a global client-aware prototype bank that captures clinically meaningful modality priors across institutions. Our Mixture of Experts is conditioned on these prototypes and modality indices to enable direction-aware expert routing for dynamically synthesizing missing features. We perform extensive quantitative and qualitative evaluations on four public chest X-ray datasets (MIMIC-CXR, NIH Open-I, PadChest, and CheXpert) and demonstrate that ProMoE-FL consistently outperforms state-of-the-art methods in both homogeneous as well as the more challenging heterogeneous settings.

Aavash Chhetri, Bibek Niroula, Eduard Vazquez +4

NepAl Applied Mathematics and Informatics Institute for research, Nepal · Fogsphere (Redev.AI Ltd), UK · University of Lausanne, Switzerland +4

Multimodal Federated Learning Missing-Modality Learning Medical Imaging Prototype Learning Mixture-of-Experts Models Federated Learning

Jul 7, 2026cs.DC

UBEP: Re-architecting Expert Parallelism Communication Library for Production Superpods

The deployment of Mixture-of-Experts (MoE) models on production high-bandwidth superpods, such as NVIDIA's NVL72/576 and Huawei's CloudMatrix384, introduces critical challenges beyond raw interconnect bandwidth. While these systems provide unified global address spaces and high-bandwidth fabrics, their full potential for sparse MoE communication is hindered by three fundamental bottlenecks: (1) Strict execution serialization imposed by coarse-grained Bulk Synchronous Parallel (BSP) orchestration of interdependent communication phases; (2) Prohibitive synchronization overhead that fails to scale alongside high interconnect bandwidth; and (3) Severe load imbalance resulting from distance-agnostic scheduling of irregular token traffic. To eliminate these bottlenecks, we introduce UBEP (Unified-Bus Expert Parallelism), a production-ready communication library that rethinks MoE's All-to-All primitives for modern superpod architectures. Through large scale experiments, UBEP reduces All-to-All latency by up to 52.4% and MoE inference Time Per Output Token (TPOT) by up to 11.1%.

Yipeng Liu, Chang Liu, Si Shen +16

State Key Laboratory for Novel Software Technology, Nanjing University · Huawei Technologies Co., Ltd.

Expert Parallelism High-Performance Computing Mixture-of-Experts Inference Mixture-of-Experts Models

Jul 7, 2026cs.CV

RoME: Robust Mixture of Low-Rank Experts against Multiple Adversarial Perturbations

Multi-perturbation adversarial training (MAT) aims to achieve robustness against multiple

\ell_p

perturbations but suffers from robustness trade-offs between different threats. To address this, we employ a mixture of experts (MoE) to route different threats through distinct model pathways. However, naive application of MoE encounters two critical challenges: experts tend to overlook threat-specific features and redundantly capture features shared across threats, and gating networks suffer from threat-agnostic routing where they learn nearly identical routing patterns across threats, thus preventing the construction of threat-specific model pathways. To this end, we propose Robust Mixture of Low-Rank Experts (RoME), where each expert is a low-rank additive update to the shared backbone, allowing it to capture threat-common features while experts focus on threat-specific information. To address threat-agnostic routing, RoME introduces (i) dual-scale gating that exploits threat-discriminative signals from local and global level features, and (ii) threat-guided gating diversification that enforces diverse expert utilization across threats. Extensive experiments demonstrate that RoME outperforms existing state-of-the-art MAT in union robustness and natural accuracy and improves robustness against unseen threats. Codes are available at https://github.com/wkim97/RoME.

Woo Jae Kim, Kyle Min, Suhyeon Ha +2

KAIST, Daejeon, Korea · Oracle, Seattle WA, USA

Adversarial Training Neural Network Robustness Adversarial Robustness Mixture-of-Experts Models

Jul 7, 2026cs.LG

NEST: Tackling Dataset-Level Distribution Shifts via Regime-Oriented Mixture-of-Experts

Accurate long-term forecasting in complex systems is frequently compromised by dataset-level distribution shifts, where diverse underlying behavioral modes and evolving system states drive the dynamic multivariate time-series. While existing methods predominantly focus on local temporal shifts, they fail to explicitly model the global structural challenge where datasets are composites of distinct operational regimes. In this paper, we propose NEST, a specialized framework designed to model and recompose these evolving structures through a two-phase dense MoE architecture. NEST first facilitates structural specialization by partitioning the dataset into distinct operational regimes through unsupervised clustering in a principled moment-entropy space. We introduce a regime-oriented router mechanism that generates initial expert weights based on temporal content, subsequently refined through geometric modulation to regime centroids. Crucially, rather than acting as monolithic predictors, individual experts function as specialized kernels that capture regime-specific dynamics by evolving unique variate-attention patterns. Extensive evaluations on diverse benchmarks, including heterogeneous network traffic and physical phenomena, demonstrate that NEST consistently achieves state-of-the-art performance. Our code and datasets are available at https://github.com/Aaralshin/NEST

Lanhao Li, Bingshu Xie, Lijun Sun +3

1Beihang University

Multivariate Time Series Forecasting Long-Term Time Series Forecasting Time Series Forecasting OOD Generalization Mixture-of-Experts Models Time Series Clustering

Jul 6, 2026cs.AI

MoP-JEPA: Hard-Assigned Predictor Mixtures for Stochastic JEPA World Models

JEPA world models commonly predict the next latent state with one regressor. Under stochastic transitions, squared and cosine regression return the conditional mean and its normalized direction, respectively: a single compromise that may match no valid successor. MoP-JEPA instead uses

K

hard-assigned heads and a context-only router to produce a finite candidate set in one pass. On held-out OGBench transitions, graph search with single-output predictors succeeds on

0.02

--

0.09

of queries, whereas MoP-JEPA reaches

0.85

. To distinguish useful successors from indiscriminate coverage, we also measure verified-route success (\emph{realroute}), which checks after graph construction whether the proposal contains a path of real transitions. MoP-JEPA leads this same-protocol metric on all three mazes; an MDN attains high raw coverage but predicts many nonexistent edges.

Zhi Song, Ximing Xing, Zhenchao Tang +8

City University of Hong Kong, China · Tencent, China

World Model Learning Latent Dynamics Modeling World Model-Based Planning Joint-Embedding Predictive Architecture Mixture-of-Experts Models World Model Evaluation

Jul 6, 2026cs.LG

Localized LoRA-MoE: Block-wise Low-Rank Experts With Adaptive Routing

Large Language Models (LLMs) and high-dimensional perception networks increasingly rely on parameter-efficient fine-tuning (PEFT) to adapt to diverse operational contexts. However, standard methods like LoRA are structurally limited by a monolithic bottleneck, making them highly susceptible to gradient warfare. Interleaved multi-task streams may trigger destructive optimization feedback, collapsing adapter weights into unspecialized averages. While recent spatial partitioning methods have introduced block-wise isolation, they remain trapped in static topologies, unable to adapt to dynamic task-switching or environmental sensor failure. In this work, we introduce Localized LoRA-MoE, a unified framework that fuses localized spatial blocking with dynamic, context-conditioned routing. We propose and evaluate two novel architectural paradigms: Block-Wise LoRA-MoE (Centralized Macro-Routing), which modulates the entire structural grid via a monolithic context signal, and Cell-Wise LoRA-MoE (Decentralized Micro-Routing), which empowers every coordinate cell in the matrix grid with autonomous, localized expert gating. Through a comprehensive suite of benchmarks, ranging from high-dimensional SVD matrix simulations and real-world tabular transformations to spatial vision perception under sensor degradation, we demonstrate that both architectures resolve optimization deadlocks inherent in static baselines. Our empirical results establish that decentralized cell-level gating achieves complete statistical parity with an omniscient global coordinator, providing a robust "gradient firewall" that protects surviving pathways from fault-propagated corruption. Our proposals consistently outperform static baselines, offering a scalable and parameter-efficient solution for dynamic model adaptation across granular coordinate fields and shifting operational regimes.

Babak Barazandeh, Subhabrata Majumdar, Vinay Prithyani +1

Fortinet · Indian Institute of Management Bangalore · Citadel Securities +1

Adaptive Model Routing Low-Rank Adaptation Mixture-of-Experts Models Parameter-Efficient Fine-Tuning

Jul 5, 2026cs.AI

Nemotron-Labs-3-Puzzle-75B-A9B: Compressing Hybrid MoE LLMs

We present Nemotron-Labs-3-Puzzle-75B-A9B, a compressed variant of Nemotron-3-Super optimized for interactive deployment. We designed the model to maximize server throughput under high user throughput constraints. In interactive serving workloads on a single 8xB200 node, Puzzle-75B-A9B achieves approximately 2x higher server throughput than Nemotron-3-Super at matched user throughput constraints. In ultra-long-context deployment on a single H100 GPU, the compressed model increases 1M-token concurrency from 1 request to 8 requests. Puzzle-75B-A9B is constructed using a multi-stage pipeline that combines the Iterative Puzzle compression framework with knowledge distillation, reinforcement learning, quantization, and a Multi-Token Prediction head. The compression process jointly optimizes heterogeneous MoE pruning, active parameter budget, and Mamba pruning to improve inference efficiency while preserving model quality. We evaluate Puzzle-75B-A9B on a broad suite of reasoning, coding, multilingual, long-context, and agentic benchmarks. Despite substantial compression, the model retains strong downstream accuracy relative to the parent model across a wide range of tasks. These results demonstrate that large hybrid MoE models can be substantially optimized for deployment efficiency while maintaining strong downstream capability. Our model is publicly available on Hugging Face.

Akhiad Bercovich, Talor Abramovich, Daniel Afrimi +67

Mixture-of-Experts Pruning LLM Pruning Mixture-of-Experts Language Models LLM Compression Long-Context Language Model Inference Mixture-of-Experts Models