| InternImage: Exploring Large-Scale Vision Foundation Models with Deformable Convolutions | ✓ Link | 71.7% | InternImage-H | 2022-11-10 |
| DINOv3 | ✓ Link | 70.0 | DINOv3 (7B) | 2025-08-13 |
| MixMAE: Mixed and Masked Autoencoder for Efficient Pretraining of Hierarchical Vision Transformers | ✓ Link | 69.3 | MixMIM-L | 2022-05-26 |
| Vision Models Are More Robust And Fair When Pretrained On Uncurated Images Without Supervision | ✓ Link | 69.0 | SEER (RegNet10B - finetuned - 384px) | 2022-02-16 |
| MixMAE: Mixed and Masked Autoencoder for Efficient Pretraining of Hierarchical Vision Transformers | ✓ Link | 68.3 | MixMIM-B | 2022-05-26 |
| DINOv2: Learning Robust Visual Features without Supervision | ✓ Link | 67.5 | DINOv2 (ViT-g/14) | 2023-04-14 |
| Masked Autoencoders Are Scalable Vision Learners | ✓ Link | 66.8 | MAE (ViT-H, 448) | 2021-11-11 |
| Self-supervised Pretraining of Visual Features in the Wild | ✓ Link | 66.0 | SEER | 2021-03-02 |
| Boosting Discriminative Visual Representation Learning with Scenario-Agnostic Mixup | ✓ Link | 64.3 | SAMix (ResNet-50 Supervised) | 2021-11-30 |
| AutoMix: Unveiling the Power of Mixup for Stronger Classifiers | ✓ Link | 64.1 | AutoMix (ResNet-50 Supervised) | 2021-03-24 |
| Revisiting Feature Prediction for Learning Visual Representations from Video | ✓ Link | 62.8 | V-JEPA ViT-H/16 (384) | 2024-02-15 |
| Self-supervised Pretraining of Visual Features in the Wild | ✓ Link | 62.7 | RegNetY-128GF (Supervised) | 2021-03-02 |
| Self-Supervised Learning from Images with a Joint-Embedding Predictive Architecture | ✓ Link | 59.1 | I-JEPA ViT-G/16 | 2023-01-19 |
| Unsupervised Learning of Visual Features by Contrasting Cluster Assignments | ✓ Link | 56.7% | SwAV | 2020-06-17 |
| Barlow Twins: Self-Supervised Learning via Redundancy Reduction | ✓ Link | 54.1% | Barlow Twins (ResNet-50) | 2021-03-04 |
| Bootstrap your own latent: A new approach to self-supervised Learning | ✓ Link | 54.0 | BYOL | 2020-06-13 |
| A Simple Framework for Contrastive Learning of Visual Representations | ✓ Link | 53.3 | SimCLR | 2020-02-13 |
| Unsupervised Learning of Visual Features by Contrasting Cluster Assignments | ✓ Link | 53.2% | ResNet-50 (Supervised) | 2020-06-17 |
| SiamMM: A Mixture Model Perspective on Deep Unsupervised Learning | | 53.2 | SiamMM | 2025-11-07 |
| Improved Baselines with Momentum Contrastive Learning | ✓ Link | 52.9 | MoCo v2 | 2020-03-09 |