paper-with-me

Papers

SMoES: Soft Modality-Guided Expert Specialization in MoE-VLMs

2026-04-27 · Zi-Hao Bo, Yaqian Li, Anzhou Hou, Rinyoichi Takezoe, Ertao Zhao, Tianxiang Pan, Jiale Yan, Mo Guang, Kaiwen Long arxiv

Mixture-of-Experts (MoE) has become a prevalent backbone for large vision-language models (VLMs), yet how modality-specific signals should guide expert routing remains under-explored. Existing routing strategies are either hand-crafted or modality-agnostic, relying on idealized priors that ignore the layer-dependent modality fusion patterns in MoE-VLMs and provide little guidance for expert specialization. We propose Soft Modality-guided Expert Specialization (SMoES), which consists of dynamic soft modality scores that capture layer-dependent fusion patterns, an expert binning mechanism aligned with expert-parallel deployment, and an inter-bin mutual information regularization that encourages coherent modality specialization. Our method leverages attention-based or Gaussian-statistics modality scores to optimize mutual information regularization. Experiments across four MoE-based VLMs and 16 benchmarks demonstrate improvement on both effectiveness and efficiency: 0.9% and 4.2% average gain on multimodal and language-only tasks, 56.1% reduction in EP communication overhead, and 12.3% throughput improvement under realistic deployment. These results validate that aligning routing with modality-aware expert specialization unlocks MoE-VLM capacity and efficiency.

📄 PDF Abstract BibTeX arXiv:2604.23996

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

H3M-SSMoEs: Hypergraph-based Multimodal Learning with LLM Reasoning and Style-Structured Mixture of Experts

2025-10-29 · Peilin Tan, Liang Xie, Churan Zhi, Dian Tu 외 arxiv

Stock movement prediction remains fundamentally challenging due to complex temporal dependencies, heterogeneous modalities, and dynamically evolving inter-stock relationships. Existing approaches often fail to unify stru…

MedMix: Specialization-Consistent Federated Sparse MoEs under Modality Heterogeneity

2026-08-14 · Adiba Orzikulova, Dong Min Kim, Jaehong Yoon, Sung-Ju Lee arxiv

Federated multimodal medical AI faces modality heterogeneity at both the client and sample levels: clients may systematically lack access to specific modality types, while individual records within the same client may co…

Federated Learning

Sparse MoE as the New Dropout: Scaling Dense and Self-Slimmable Transformers

2023-03-02 · Tianlong Chen, Zhenyu Zhang, Ajay Jaiswal, Shiwei Liu 외

Despite their remarkable achievement, gigantic transformers encounter significant drawbacks, including exorbitant computational and memory footprints during training, as well as severe collapse evidenced by a high degree…

Mixture-of-Experts

Finding Fantastic Experts in MoEs: A Unified Study for Expert Dropping Strategies and Observations

2025-04-08 · Ajay Jaiswal, Jianyu Wang, Yixiao Li, Pingzhi Li 외

Sparsely activated Mixture-of-Experts (SMoE) has shown promise in scaling up the learning capacity of neural networks. However, vanilla SMoEs have issues such as expert redundancy and heavy memory requirements, making th…

Instruction FollowingMixture-of-Experts

CoSMoEs: Compact Sparse Mixture of Experts

2025-02-28 · Patrick Huber, Akshat Shrivastava, Ernie Chang, Chinnadhurai Sankar 외

Sparse Mixture of Expert (MoE) models are popular foundational architectures at large scale, however, under-explored at smaller sizes. Here, we show how to enable Compact Sparse Mixture of Experts (CoSMoEs) for on-device…

Mixture-of-Experts