Loading the SOTA2 catalog…
SOTA2 Research · benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
| Task | Dataset | Domain | Trend | Results | Last Update |
|---|---|---|---|---|---|
| Image Classification | CIFAR-LT-100 (test) | 74 | Feb 26, 2026 | ||
| 3D Human Pose and Shape Estimation | 3DPW | 74 | Apr 28, 2026 | ||
| Counting | CARPK | 74 | Jul 3, 2026 | ||
| Video Captioning | VATEX (test) | 74 | Jul 9, 2026 | ||
| Image Retrieval | Revisited Paris (RPar) + R1M (Medium) |
| 74 |
| Feb 26, 2026 |
| Robustness to Corruptions | ImageNet-C (test) | 74 | Mar 18, 2026 |
|---|
| Out-of-Distribution Detection | ImageNet-O | 74 | Feb 26, 2026 |
|---|
| Action Recognition | PKU-MMD Part I | 74 | Mar 11, 2026 |
|---|
| Anomaly Segmentation | Fishyscapes Lost & Found (val) | 74 | Mar 5, 2026 |
|---|
| Offline Reinforcement Learning | D4RL antmaze-umaze (diverse) | 74 | May 29, 2026 |
|---|
| Domain Adaptation | Office31 (test) | 74 | Feb 26, 2026 |
|---|
| Salient Object Detection | DUTS | 74 | Jul 1, 2026 |
|---|
| OOD Detection | CIFAR-10 IND LSUN C OOD | 74 | Jul 8, 2026 |
|---|
| Face forgery detection | DFDC | 74 | May 29, 2026 |
|---|
| Image Denoising | Set5 | 74 | May 14, 2026 |
|---|
| Action Recognition | Kinetics 700 | 74 | Mar 17, 2026 |
|---|
| Image Generation | ImageNet 256x256 (train val) | 74 | Jun 30, 2026 |
|---|
| Visual Question Answering | GQA (test-std) | 74 | May 7, 2026 |
|---|
| Natural Language Understanding | SuperGLUE (test) | 74 | May 21, 2026 |
|---|
| Video-to-Text Retrieval | MSR-VTT (1k-A) | 74 | Feb 27, 2026 |
|---|
| Image Classification | Dogs vs. Cats (test) | 74 | May 27, 2026 |
|---|
| Monocular depth estimation | Cityscapes | 74 | May 27, 2026 |
|---|
| 3D Reconstruction | ShapeNet (test) | 74 | Feb 26, 2026 |
|---|
| Image Geolocalization | YFCC4k | 74 | Jul 7, 2026 |
|---|
| Semantic Segmentation | Cityscapes v1 (test) | 74 | Feb 26, 2026 |
|---|