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 |
|---|---|---|---|---|---|
| Continuous Robot Control | HalfCheetah v3 (test) | 48 | Feb 26, 2026 | ||
| News Classification | AG News (test) | 48 | May 12, 2026 | ||
| Image Classification | ImageNet 1K 1.0 (val) | 48 | Feb 26, 2026 | ||
| Argument Identification and Classification | ACE05-E (test) | 48 | Feb 26, 2026 | ||
| Argument Identification and Classification | ACE05-E (dev) | 48 | Feb 26, 2026 | ||
| Argument Classification |
|---|
| ACE05-E (dev) |
| 48 |
| Feb 26, 2026 |
| Named Entity Recognition | bc2gm | 48 | Apr 20, 2026 |
|---|
| Classification | YaleB (test) | 48 | Feb 26, 2026 |
|---|
| Denoising | Spaces frame 6 (test) | 48 | Feb 26, 2026 |
|---|
| Novel View Synthesis | LLFF forward-facing 27 | 48 | Feb 26, 2026 |
|---|
| Crowd Counting | RGBT-CC (test) | 48 | May 19, 2026 |
|---|
| Image Classification | SUN397 base-to-new generalization | 48 | Feb 26, 2026 |
|---|
| Person Re-identification | Market-1501 -> MSMT17 (M→MS) (test) | 48 | Jun 16, 2026 |
|---|
| Image Classification | Tiny-Imagenet Resnet20 (test) | 48 | Feb 26, 2026 |
|---|
| Pulse rate estimation | PURE Intra-dataset | 48 | Jun 11, 2026 |
|---|
| Receptor Docking Affinity | TDC DRD3 (leaderboard) | 48 | Feb 26, 2026 |
|---|
| Blind Image Quality Assessment | CLIVE | 48 | Feb 26, 2026 |
|---|
| Node Classification | penn94 (test) | 48 | May 20, 2026 |
|---|
| Panoptic Segmentation | Cityscapes | 48 | May 20, 2026 |
|---|
| Sentence Completion | COPA | 48 | Feb 26, 2026 |
|---|
| Semantic Segmentation | COCO 20^i (test) | 48 | Feb 26, 2026 |
|---|
| Image Classification | ImageNet-50 (test) | 48 | May 26, 2026 |
|---|
| Node Classification | Cora | 48 | Feb 27, 2026 |
|---|
| No-Reference Image Quality Assessment | IQA inter-model attack dataset | 48 | Feb 26, 2026 |
|---|
| Attribution Fidelity | ImageNet 1,000 images (val) | 48 | Feb 26, 2026 |
|---|