Distilled Large Language Model-Driven Dynamic Sparse Expert Activation Mechanism

• URL: http://arxiv.org/abs/2603.26735v1
• Date: Sat, 21 Mar 2026 08:46:03 GMT
• Title: Distilled Large Language Model-Driven Dynamic Sparse Expert Activation Mechanism
• Abstract summary: Distilled Large Language Model (LLM)-Driven Sparse Mixture-of-Experts (DS-MoE) framework integrates text-guided dynamic routing and lightweight multi-scale comprehension.<n>textbfDS-MoE surpasses YOLOv8/YOLOX with gains of +13.9, +1.4, and +2.0 pp mAP@ 0.5:0.95 on BBMP, aluminum, and PCB, respectively.
• Score: 28.84340005858496
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/
• Abstract: High inter-class similarity, extreme scale variation, and limited computational budgets hinder reliable visual recognition across diverse real-world data. Existing vision-centric and cross-modal approaches often rely on rigid fusion mechanisms and heavy annotation pipelines, leading to sub-optimal generalization. We propose the Distilled Large Language Model (LLM)-Driven Sparse Mixture-of-Experts (DS-MoE) framework, which integrates text-guided dynamic routing and lightweight multi-scale comprehension. The DS-MoE framework dynamically aligns textual semantics with defect-specific visual patterns through a sparse MoE architecture, where task-relevant experts are adaptively activated based on semantic relevance, resolving inter-class ambiguity. A lightweight MobileSAM encoder enables real-time inference while preserving multi-scale defect details. Extensive experiments on PCB, aluminum foil, and mold defect datasets demonstrate that our framework achieves superior performance compared to existing pure vision models. \textbf{DS-MoE} surpasses YOLOv8/YOLOX with gains of +13.9, +1.4, and +2.0 pp mAP@ 0.5:0.95 on BBMP, aluminum, and PCB, respectively, while also improving precision and recall.

Related papers

• InternVL-U: Democratizing Unified Multimodal Models for Understanding, Reasoning, Generation and Editing [87.5130783098133]
  InternVL-U is a lightweight 4B- parameter UMM that democratizes multimodal understanding and reasoning capabilities.<n>It integrates a state-of-the-art Multimodal Large Language Model (MLLM) with a specialized MMDiT-based visual generation head.<n>It consistently outperforms unified baseline models with over 3x larger scales such as BAGEL (14B) on various generation and editing tasks.
  arXiv  Detail & Related papers  (2026-03-10T16:38:33Z)
• Beyond Language Modeling: An Exploration of Multimodal Pretraining [125.34714978184638]
  We provide empirical clarity through controlled, from-scratch pretraining experiments.<n>We adopt the Transfusion framework, using next-token prediction for language and diffusion for vision.<n>We demonstrate that the MoE architecture harmonizes this scaling asymmetry by providing the high model capacity required by language.
  arXiv  Detail & Related papers  (2026-03-03T18:58:00Z)
• Mixture-of-Experts Models in Vision: Routing, Optimization, and Generalization [0.0]
  We study MoE behavior in an image classification setting, focusing on predictive performance, expert utilization, and generalization.<n>We compare dense, SoftMoE, and SparseMoE classifier heads on the CIFAR10 dataset under comparable model capacity.<n>Both MoE variants achieve slightly higher validation accuracy than the dense baseline while maintaining balanced expert utilization through regularization.<n>We find that SoftMoE exhibits higher sharpness by these metrics, while Dense and SparseMoE lie in a similar curvature regime, despite all models achieving comparable generalization performance.
  arXiv  Detail & Related papers  (2026-01-21T14:22:25Z)
• Multi-Modal Interpretability for Enhanced Localization in Vision-Language Models [2.984679075401059]
  This paper presents the Multi-Modal Explainable Learning framework, designed to enhance the interpretability of vision-language models.<n>Our approach processes features at multiple semantic levels to capture relationships between image regions at different granularities.<n>We show that by incorporating semantic relationship information into gradient-based attribution maps, MMEL produces more focused and contextually aware visualizations.
  arXiv  Detail & Related papers  (2025-09-17T18:18:59Z)
• OneCAT: Decoder-Only Auto-Regressive Model for Unified Understanding and Generation [91.45421429922506]
  OneCAT is a unified multimodal model that seamlessly integrates understanding, generation, and editing.<n>Our framework eliminates the need for external components such as Vision Transformers (ViT) or vision tokenizer during inference.
  arXiv  Detail & Related papers  (2025-09-03T17:29:50Z)
• FUDOKI: Discrete Flow-based Unified Understanding and Generation via Kinetic-Optimal Velocities [76.46448367752944]
  multimodal large language models (MLLMs) unify visual understanding and image generation within a single framework.<n>Most existing MLLMs rely on autore (AR) architectures, which impose inherent limitations on future development.<n>We introduce FUDOKI, a unified multimodal model purely based on discrete flow matching.
  arXiv  Detail & Related papers  (2025-05-26T15:46:53Z)
• EMMA: Efficient Visual Alignment in Multi-Modal LLMs [56.03417732498859]
  EMMA is a lightweight cross-modality module designed to efficiently fuse visual and textual encodings.<n>EMMA boosts performance across multiple tasks by up to 9.3% while significantly improving robustness against hallucinations.
  arXiv  Detail & Related papers  (2024-10-02T23:00:31Z)
• DualKanbaFormer: An Efficient Selective Sparse Framework for Multimodal Aspect-based Sentiment Analysis [0.6187939267100836]
  We introduce DualKanbaFormer, a novel framework that leverages parallel Textual and Visual KanbaFormer modules for robust multimodal analysis.<n>Our approach incorporates Aspect-Driven Sparse Attention (ADSA) to balance coarse-grained aggregation and fine-grained selection for aspect-focused precision.<n>We replace traditional feed-forward networks and normalization with Kolmogorov-Arnold Networks (KANs) and Dynamic Tanh (DyT) to enhance non-linear expressivity and inference stability.
  arXiv  Detail & Related papers  (2024-08-27T19:33:15Z)
• Scaling Vision-Language Models with Sparse Mixture of Experts [128.0882767889029]
  We show that mixture-of-experts (MoE) techniques can achieve state-of-the-art performance on a range of benchmarks over dense models of equivalent computational cost. Our research offers valuable insights into stabilizing the training of MoE models, understanding the impact of MoE on model interpretability, and balancing the trade-offs between compute performance when scaling vision-language models.
  arXiv  Detail & Related papers  (2023-03-13T16:00:31Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.