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 |
|---|---|---|---|---|---|
| Class Incremental Panoptic Segmentation | ADE20K (val) | 32 | Feb 26, 2026 | ||
| Instance Segmentation | COCO | 32 | Feb 26, 2026 | ||
| Mathematical Reasoning | MATH (test) | 32 | Feb 26, 2026 | ||
| Semantic Segmentation | ScanNet++ (val) | 32 | Mar 4, 2026 | ||
| WSI classification | PANDA | 32 | May 19, 2026 | ||
| Scene Graph Anticipation |
|---|
| Action Genome AGS, F=0.5 |
| 32 |
| Feb 26, 2026 |
| Anomaly Detection | Liver CT | 32 | Jun 30, 2026 |
|---|
| Class-Incremental Learning | CUB200 (100-20) | 32 | May 14, 2026 |
|---|
| Instruction Following | ComplexInstruct Level 2 | 32 | Feb 26, 2026 |
|---|
| General Language Understanding | Standard Downstream Tasks Suite (SciQ, PIQA, WinoGrande, ARC-E, ARC-C, HellaSwag, LogiQA, BoolQ, LAMBADA, MMLU) | 32 | Feb 26, 2026 |
|---|
| Novel view synthesis | CO3D | 32 | Jul 1, 2026 |
|---|
| Safety Evaluation | XSTest | 32 | Jun 1, 2026 |
|---|
| Identity Protection | VGG Face (test) | 32 | Feb 26, 2026 |
|---|
| Traffic Forecasting | SD | 32 | Jun 30, 2026 |
|---|
| Graph Anomaly Detection | Amazon (test) | 32 | Feb 26, 2026 |
|---|
| Graph Anomaly Detection | BlogCatalog (test) | 32 | May 27, 2026 |
|---|
| Graph Anomaly Detection | ACM (test) | 32 | Feb 26, 2026 |
|---|
| Graph Anomaly Detection | Cora (test) | 32 | Feb 26, 2026 |
|---|
| Random Inpainting | FFHQ 256x256-1K | 32 | Apr 29, 2026 |
|---|
| PAN-sharpening | WV3 Reduced-Resolution | 32 | Feb 26, 2026 |
|---|
| Image Anomaly Detection | MVTec AD | 32 | Feb 26, 2026 |
|---|
| Node classification | Telegram | 32 | Mar 4, 2026 |
|---|
| Color Image JPEG Compression Artifact Removal | Urban100 color (test) | 32 | Feb 26, 2026 |
|---|
| Color Image JPEG Compression Artifact Removal | LIVE1 color (test) | 32 | Feb 26, 2026 |
|---|
| Grayscale image JPEG compression artifact removal | Urban100 | 32 | Feb 26, 2026 |
|---|