vGamba: Attentive State Space Bottleneck for efficient Long-range Dependencies in Visual Recognition
Abstract
vGamba, a hybrid vision backbone merging state-space models with attention mechanisms, achieves efficient and accurate long-range dependency modeling in vision tasks.
Capturing long-range dependencies efficiently is essential for visual recognition tasks, yet existing methods face limitations. Convolutional neural networks (CNNs) struggle with restricted receptive fields, while Vision Transformers (ViTs) achieve global context and long-range modeling at a high computational cost. State-space models (SSMs) offer an alternative, but their application in vision remains underexplored. This work introduces vGamba, a hybrid vision backbone that integrates SSMs with attention mechanisms to enhance efficiency and expressiveness. At its core, the Gamba bottleneck block that includes, Gamba Cell, an adaptation of Mamba for 2D spatial structures, alongside a Multi-Head Self-Attention (MHSA) mechanism and a Gated Fusion Module for effective feature representation. The interplay of these components ensures that vGamba leverages the low computational demands of SSMs while maintaining the accuracy of attention mechanisms for modeling long-range dependencies in vision tasks. Additionally, the Fusion module enables seamless interaction between these components. Extensive experiments on classification, detection, and segmentation tasks demonstrate that vGamba achieves a superior trade-off between accuracy and computational efficiency, outperforming several existing models.
Community
This is an automated message from the Librarian Bot. I found the following papers similar to this paper.
The following papers were recommended by the Semantic Scholar API
- ECViT: Efficient Convolutional Vision Transformer with Local-Attention and Multi-scale Stages (2025)
- Small Vision-Language Models: A Survey on Compact Architectures and Techniques (2025)
- MAAM: A Lightweight Multi-Agent Aggregation Module for Efficient Image Classification Based on the MindSpore Framework (2025)
- LSNet: See Large, Focus Small (2025)
- GFT: Gradient Focal Transformer (2025)
- DVHGNN: Multi-Scale Dilated Vision HGNN for Efficient Vision Recognition (2025)
- Cross Paradigm Representation and Alignment Transformer for Image Deraining (2025)
Please give a thumbs up to this comment if you found it helpful!
If you want recommendations for any Paper on Hugging Face checkout this Space
You can directly ask Librarian Bot for paper recommendations by tagging it in a comment:
@librarian-bot
recommend
Models citing this paper 0
No model linking this paper
Datasets citing this paper 0
No dataset linking this paper
Spaces citing this paper 0
No Space linking this paper
Collections including this paper 0
No Collection including this paper