mlsanityData to 7 Oct 2026

MambaVision: A Hybrid Mamba-Transformer Vision Backbone

arXiv 2407.08083 ↗First posted 10 Jul 2024cs.CVcited by 222

On the map: State-space sequence models (S4, Mamba) → Vision Mamba backbones

Abstract (arXiv)

We propose a novel hybrid Mamba-Transformer backbone, MambaVision, specifically tailored for vision applications. Our core contribution includes redesigning the Mamba formulation to enhance its capability for efficient modeling of visual features. Through a comprehensive ablation study, we demonstrate the feasibility of integrating Vision Transformers (ViT) with Mamba. Our results show that equipping the Mamba architecture with self-attention blocks in the final layers greatly improves its capacity to capture long-range spatial dependencies. Based on these findings, we introduce a family of MambaVision models with a hierarchical architecture to meet various design criteria. For classification on the ImageNet-1K dataset, MambaVision variants achieve state-of-the-art (SOTA) performance in terms of both Top-1 accuracy and throughput. In downstream tasks such as object detection, instance segmentation, and semantic segmentation on MS COCO and ADE20K datasets, MambaVision outperforms comparably sized backbones while demonstrating favorable performance. Code: https://github.com/NVlabs/MambaVision

Key facts. Cited by 222 papers in the map (data through 7 Oct 2026). Placed in Vision Mamba backbones, one of 10 directions in State-space sequence models (S4, Mamba). Builds on 39 of its 44 references that are in the map. First posted on arXiv 10 Jul 2024.

Builds on

39 of 44 references.

Closest ancestors · 3

Same direction, newest first

  1. 2024SiMBA: Simplified Mamba-Based Architecture for Vision and Multivariate Time series22 Mar
  2. 2024EfficientVMamba: Atrous Selective Scan for Light Weight Visual Mamba15 Mar
  3. 2024Vision Mamba: Efficient Visual Representation Learning with Bidirectional State Space Model17 Jan

Other references · 36

  1. 2017Attention Is All You Needcited by 59,932
  2. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  3. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 29,915
  4. 2014Microsoft COCO: Common Objects in Contextcited by 17,498
  5. 2021Swin Transformer: Hierarchical Vision Transformer using Shifted Windowscited by 12,082
  6. 2017Mask R-CNNcited by 7,259
  7. 2015Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shiftcited by 6,979
  8. 2023Mamba: Linear-Time Sequence Modeling with Selective State Spacescited by 4,616
  9. 2020Training data-efficient image transformers & distillation through attentioncited by 4,560
  10. 2022A ConvNet for the 2020scited by 3,545
  11. 2016Gaussian Error Linear Units (GELUs)cited by 3,431
  12. 2016Aggregated Residual Transformations for Deep Neural Networkscited by 2,540
24 more
  1. 2021Pyramid Vision Transformer: A Versatile Backbone for Dense Prediction without Convolutionscited by 1,960
  2. 2018Unified Perceptual Parsing for Scene Understandingcited by 1,265
  3. 2021Swin Transformer V2: Scaling Up Capacity and Resolutioncited by 1,215
  4. 2021EfficientNetV2: Smaller Models and Faster Trainingcited by 1,127
  5. 2024VMamba: Visual State Space Modelcited by 955
  6. 2020Designing Network Design Spacescited by 954
  7. 2017Sigmoid-Weighted Linear Units for Neural Network Function Approximation in Reinforcement Learningcited by 935
  8. 2021CrossViT: Cross-Attention Multi-Scale Vision Transformer for Image Classificationcited by 806
  9. 2019Large Batch Optimization for Deep Learning: Training BERT in 76 minutescited by 671
  10. 2021Transformer in Transformercited by 632
  11. 2021Combining Recurrent, Convolutional, and Continuous-time Models with Linear State-Space Layerscited by 607
  12. 2021Twins: Revisiting the Design of Spatial Attention in Vision Transformerscited by 565
  13. 2021MetaFormer Is Actually What You Need for Visioncited by 523
  14. 2022MaxViT: Multi-Axis Vision Transformercited by 476
  15. 2021LeViT: a Vision Transformer in ConvNet's Clothing for Faster Inferencecited by 475
  16. 2022DeiT III: Revenge of the ViTcited by 417
  17. 2021ResNet strikes back: An improved training procedure in timmcited by 365
  18. 2021XCiT: Cross-Covariance Image Transformerscited by 335
  19. 2022EfficientFormer: Vision Transformers at MobileNet Speedcited by 240
  20. 2021Co-Scale Conv-Attentional Image Transformerscited by 224
  21. 2021Visformer: The Vision-friendly Transformercited by 137
  22. 2022Global Context Vision Transformerscited by 98
  23. 2022Next-ViT: Next Generation Vision Transformer for Efficient Deployment in Realistic Industrial Scenarioscited by 65
  24. 2023FasterViT: Fast Vision Transformers with Hierarchical Attentioncited by 65

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 29; 2025: 113; 2026: 80.

222 papers.

Sort:

By area:

  1. 2024Transformers to SSMs: Distilling Quadratic Knowledge to Subquadratic Modelscited by 66
  2. 2024A Survey of Mambacited by 39
  3. 2025Scaling Laws in Patchification: An Image Is Worth 50,176 Tokens And Morecited by 28
  4. 2025Mamba as a Bridge: Where Vision Foundation Models Meet Vision Language Models for Domain-Generalized Semantic Segmentationcited by 23
  5. 2024Mamba Policy: Towards Efficient 3D Diffusion Policy with Hybrid Selective State Modelscited by 22
  6. 2024Mamba in Vision: A Comprehensive Survey of Techniques and Applicationscited by 18
  7. 2025MambaIC: State Space Models for High-Performance Learned Image Compressioncited by 18
  8. 2024MobileMamba: Lightweight Multi-Receptive Visual Mamba Networkcited by 16
  9. 2024MaDiNet: Mamba Diffusion Network for SAR Target Detectioncited by 13
  10. 2024MAP: Unleashing Hybrid Mamba-Transformer Vision Backbone's Potential with Masked Autoregressive Pretrainingcited by 12
  11. 2024MambaPEFT: Exploring Parameter-Efficient Fine-Tuning for Mambacited by 11
  12. 2024Mamba-SEUNet: Mamba UNet for Monaural Speech Enhancementcited by 10
38 more
  1. 2025EAMamba: Efficient All-Around Vision State Space Model for Image Restorationcited by 10
  2. 2024Adventurer: Optimizing Vision Mamba Architecture Designs for Efficiencycited by 9
  3. 2025MV-SSM: Multi-View State Space Modeling for 3D Human Pose Estimationcited by 9
  4. 2025Gated KalmaNet: A Fading Memory Layer Through Test-Time Ridge Regressioncited by 9
  5. 2025Semi-Supervised State-Space Model with Dynamic Stacking Filter for Real-World Video Derainingcited by 8
  6. 2024MaskMamba: A Hybrid Mamba-Transformer Model for Masked Image Generationcited by 7
  7. 2025TransMamba: A Sequence-Level Hybrid Transformer-Mamba Language Modelcited by 7
  8. 2025Mamba-Sea: A Mamba-based Framework with Global-to-Local Sequence Augmentation for Generalizable Medical Image Segmentationcited by 7
  9. 2025Catch-Only-One: Non-Transferable Examples for Model-Specific Authorizationcited by 7
  10. 2024HMT-UNet: A hybird Mamba-Transformer Vision UNet for Medical Image Segmentationcited by 6
  11. 2024A Hybrid Transformer-Mamba Network for Single Image Derainingcited by 6
  12. 2024MatMamba: A Matryoshka State Space Modelcited by 6
  13. 2025MaTVLM: Hybrid Mamba-Transformer for Efficient Vision-Language Modelingcited by 6
  14. 2025Dynamic Vision Mambacited by 6
  15. 2025Vision Mamba in Remote Sensing: A Comprehensive Survey of Techniques, Applications and Outlookcited by 6
  16. 2025ThinkingViT: Matryoshka Thinking Vision Transformer for Elastic Inferencecited by 6
  17. 2024Mamba-CL: Optimizing Selective State Space Model in Null Space for Continual Learningcited by 5
  18. 2024Faster Vision Mamba is Rebuilt in Minutes via Merged Token Re-trainingcited by 5
  19. 2025MambaGlue: Fast and Robust Local Feature Matching With Mambacited by 5
  20. 2025OuroMamba: A Data-Free Quantization Framework for Vision Mambacited by 5
  21. 2025ViT-Linearizer: Distilling Quadratic Knowledge into Linear-Time Vision Modelscited by 5
  22. 2025Overflow Prevention Enhances Long-Context Recurrent LLMscited by 5
  23. 2025RadioMamba: Breaking the Accuracy-Efficiency Trade-off in Radio Map Construction via a Hybrid Mamba-UNetcited by 5
  24. 2024GenMapping: Unleashing the Potential of Inverse Perspective Mapping for Robust Online HD Map Constructioncited by 4
  25. 2025Parallel Sequence Modeling via Generalized Spatial Propagation Networkcited by 4
  26. 2025Post-Training Quantization for Vision Mamba with k-Scaled Quantization and Reparameterizationcited by 4
  27. 2025Spectral State Space Model for Rotation-Invariant Visual Representation Learningcited by 4
  28. 2025Atlas: Multi-Scale Attention Improves Long Context Image Modelingcited by 4
  29. 2025An Efficient and Mixed Heterogeneous Model for Image Restorationcited by 4
  30. 2025MoMa: Modulating Mamba for Adapting Image Foundation Models to Video Recognitioncited by 4
  31. 2025MedMambaLite: Hardware-Aware Mamba for Medical Image Classificationcited by 4
  32. 2025VCMamba: Bridging Convolutions with Multi-Directional Mamba for Efficient Visual Representationcited by 4
  33. 2026MHLA: Restoring Expressivity of Linear Attention via Token-Level Multi-Headcited by 4
  34. 2026SF-Mamba: Rethinking State Space Model for Visioncited by 4
  35. 2026Diffusion Model as a Generalist Segmentation Learnercited by 4
  36. 2024Mamba-YOLO-World: Marrying YOLO-World with Mamba for Open-Vocabulary Detectioncited by 3
  37. 2024GraspMamba: A Mamba-based Language-driven Grasp Detection Framework with Hierarchical Feature Learningcited by 3
  38. 2024M3D: Dual-Stream Selective State Spaces and Depth-Driven Framework for High-Fidelity Single-View 3D Reconstructioncited by 3

50 most cited of 222 All 222 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2025A Survey on Mamba Architecture for Vision Applications
  2. 2024A Survey on Visual Mamba
  3. 2024Visual Mamba: A Survey and New Outlooks
  4. 2024MambaOut: Do We Really Need Mamba for Vision?
  5. 2024Demystify Mamba in Vision: A Linear Attention Perspective
  6. 2024Mamba-R: Vision Mamba ALSO Needs Registers
  7. 2025MatIR: A Hybrid Mamba-Transformer Image Restoration Model
  8. 2024Mamba-UNet: UNet-Like Pure Visual Mamba for Medical Image Segmentation