mlsanityData to 14 Oct 2026

DINOv3

arXiv 2508.10104 ↗First posted 13 Aug 2025cs.CV cs.LGcited by 1,414

Not on the map. Why?

Cited by 106 in Diffusion models, but placed by its own abstract, which didn’t fit.

Abstract (arXiv)

Self-supervised learning holds the promise of eliminating the need for manual data annotation, enabling models to scale effortlessly to massive datasets and larger architectures. By not being tailored to specific tasks or domains, this training paradigm has the potential to learn visual representations from diverse sources, ranging from natural to aerial images -- using a single algorithm. This technical report introduces DINOv3, a major milestone toward realizing this vision by leveraging simple yet effective strategies. First, we leverage the benefit of scaling both dataset and model size by careful data preparation, design, and optimization. Second, we introduce a new method called Gram anchoring, which effectively addresses the known yet unsolved issue of dense feature maps degrading during long training schedules. Finally, we apply post-hoc strategies that further enhance our models' flexibility with respect to resolution, model size, and alignment with text. As a result, we present a versatile vision foundation model that outperforms the specialized state of the art across a broad range of settings, without fine-tuning. DINOv3 produces high-quality dense features that achieve outstanding performance on various vision tasks, significantly surpassing previous self- and weakly-supervised foundation models. We also share the DINOv3 suite of vision models, designed to advance the state of the art on a wide spectrum of tasks and data by providing scalable solutions for diverse resource constraints and deployment scenarios.

Key facts. Cited by 1,414 papers in the map (data through 14 Oct 2026). First posted on arXiv 13 Aug 2025.

Builds on

0 of 0 references.

No references here yet.

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 221; 2026: 1193.

1,414 papers.

Sort:

By area:

  1. 2025LeJEPA: Provable and Scalable Self-Supervised Learning Without the Heuristicscited by 254
  2. 2025Native and Compact Structured Latents for 3D Generationcited by 175
  3. 2025DriveVLA-W0: World Models Amplify Data Scaling Law in Autonomous Drivingcited by 134
  4. 2026V-JEPA 2.1: Unlocking Dense Features in Video Self-Supervised Learningcited by 117
  5. 2025FastVGGT: Training-Free Acceleration of Visual Geometry Transformercited by 98
  6. 2025RF-DETR: Neural Architecture Search for Real-Time Detection Transformerscited by 75
  7. 2026SAM 3D Body: Robust Full-Body Human Mesh Recoverycited by 69
  8. 2025UniVerse-1: Unified Audio-Video Generation via Stitching of Expertscited by 61
  9. 2026LaWAM: Latent World Action Models for Efficient Dynamics-Aware Robot Policiescited by 52
  10. 2025RoMa v2: Harder Better Faster Denser Feature Matchingcited by 51
  11. 2026HY-World 2.0: A Multi-Modal World Model for Reconstructing, Generating, and Simulating 3D Worldscited by 44
  12. 2026Image Generators are Generalist Vision Learnerscited by 43
38 more
  1. 2026Temporal Straightening for Latent Planningcited by 38
  2. 2025TUNA: Taming Unified Visual Representations for Native Unified Multimodal Modelscited by 34
  3. 2026GLM-5V-Turbo: Toward a Native Foundation Model for Multimodal Agentscited by 34
  4. 2025World Models for Learning Dexterous Hand-Object Interactions from Human Videoscited by 31
  5. 2026CLAP: Contrastive Latent Action Pretraining for Learning Vision-Language-Action Models from Human Videoscited by 31
  6. 2025AnyUp: Universal Feature Upsamplingcited by 28
  7. 2025IGGT: Instance-Grounded Geometry Transformer for Semantic 3D Reconstructioncited by 28
  8. 2025Fast-FoundationStereo: Real-Time Zero-Shot Stereo Matchingcited by 28
  9. 2026OA-WAM: Object-Addressable World Action Model for Robust Robot Manipulationcited by 28
  10. 2025Real-Time Object Detection Meets DINOv3cited by 27
  11. 2025From Perception to Cognition: A Survey of Vision-Language Interactive Reasoning in Multimodal Large Language Modelscited by 26
  12. 2026VGGT-Omegacited by 26
  13. 2025Depth Any Panoramas: A Foundation Model for Panoramic Depth Estimationcited by 25
  14. 2026JEPA-VLA: Video Predictive Embedding is Needed for VLA Modelscited by 24
  15. 2026InfiniDepth: Arbitrary-Resolution and Fine-Grained Depth Estimation with Neural Implicit Fieldscited by 23
  16. 2025Comprehensive language-image pre-training for 3D medical image understandingcited by 22
  17. 2025Emergent Outlier View Rejection in Visual Geometry Grounded Transformerscited by 22
  18. 2025E-RayZer: Self-supervised 3D Reconstruction as Spatial Visual Pre-trainingcited by 22
  19. 2026D-OPSD: On-Policy Self-Distillation for Continuously Tuning Step-Distilled Diffusion Modelscited by 22
  20. 2025MoRE: 3D Visual Geometry Reconstruction Meets Mixture-of-Expertscited by 21
  21. 2025C3G: Learning Compact 3D Representations with 2K Gaussianscited by 21
  22. 2025DVGT: Driving Visual Geometry Transformercited by 21
  23. 2026Utonia: Toward One Encoder for All Point Cloudscited by 21
  24. 2025SegDINO: An Efficient Design for Medical and Natural Image Segmentation with DINO-V3cited by 20
  25. 2025UniLat3D: Geometry-Appearance Unified Latents for Single-Stage 3D Generationcited by 20
  26. 2026Diversity-Preserved Distribution Matching Distillation for Fast Visual Synthesiscited by 20
  27. 2026DualCoT-VLA: Visual-Linguistic Chain of Thought via Parallel Reasoning for Vision-Language-Action Modelscited by 20
  28. 2026EgoWAM: World Action Models Beyond Pixels with In-the-Wild Egocentric Human Datacited by 20
  29. 2025SpaceVista: All-Scale Visual Spatial Reasoning from mm to kmcited by 19
  30. 2025Does Object Binding Naturally Emerge in Large Pretrained Vision Transformers?cited by 19
  31. 2026Vision-aligned Latent Reasoning for Multi-modal Large Language Modelcited by 18
  32. 2025VFM-VAE: Vision Foundation Models Can Be Good Tokenizers for Latent Diffusion Modelscited by 17
  33. 2025GARDO: Reinforcing Diffusion Models without Reward Hackingcited by 17
  34. 2026S-VAM: Shortcut Video-Action Model by Self-Distilling Geometric and Semantic Foresightcited by 17
  35. 2025Dino U-Net: Exploiting High-Fidelity Dense Features from Foundation Models for Medical Image Segmentationcited by 16
  36. 2026Back into Plato's Cave: Examining Cross-modal Representational Convergence at Scalecited by 16
  37. 2026VISReg: Variance-Invariance-Sketching Regularization for JEPA trainingcited by 16
  38. 2025GIR-Bench: Versatile Benchmark for Generating Images with Reasoningcited by 15

50 most cited of 1,414 All 1,414 (JSON)