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 |
|---|---|---|---|---|---|
| Anomaly Classification | VisA 1.0 (test) | 34 | Feb 26, 2026 | ||
| Anomaly Segmentation | VisA 1.0 (test) | 34 | Feb 26, 2026 | ||
| Mathematical Reasoning | MATH in-domain (test) | 34 | May 13, 2026 | ||
| Image Denoising | CBSD68 sigma=15 (test) | 34 | Feb 26, 2026 | ||
| Image Registration | HCP | 34 | Feb 26, 2026 | ||
| HICO-DET |
| 34 |
| Feb 26, 2026 |
| Out-of-Distribution Detection | CLINCSMALL | 34 | Feb 26, 2026 |
|---|
| Out-of-Distribution Detection | CLINCFULL | 34 | Feb 26, 2026 |
|---|
| OOD Detection | Yelp (test) | 34 | Feb 26, 2026 |
|---|
| Skeleton-based Action Recognition | NTU-RGB+D 60 (X-Sub) | 34 | May 13, 2026 |
|---|
| 3D Face Reconstruction | REALY (frontal-view) | 34 | Feb 26, 2026 |
|---|
| Semantic Segmentation | Structured3D (test) | 34 | Mar 4, 2026 |
|---|
| Medical Image Segmentation | PROMISE12 | 34 | Jun 23, 2026 |
|---|
| Semantic Segmentation | ACDC | 34 | Jun 2, 2026 |
|---|
| Anomaly Detection | Cover | 34 | May 29, 2026 |
|---|
| Feature Interaction Attribution | Dyck-2 15,000 corpus size (test) | 34 | Feb 26, 2026 |
|---|
| Multimodal Sentiment Analysis | MOSI (test) | 34 | Feb 26, 2026 |
|---|
| Complex Query Answering | FB15k-237 (test) | 34 | Mar 18, 2026 |
|---|
| Sentiment Classification | Tweet (test) | 34 | Feb 26, 2026 |
|---|
| Sentiment Classification | Yelp5 (test) | 34 | Feb 26, 2026 |
|---|
| Sentiment Classification | Amz5 (test) | 34 | Feb 26, 2026 |
|---|
| Extractive Question Answering | SQuAD 2.0 | 34 | Feb 26, 2026 |
|---|
| 3D Scene Reconstruction | 7-scenes (test) | 34 | Jul 8, 2026 |
|---|
| Image Classification | CRC (test) | 34 | Feb 26, 2026 |
|---|
| Few-Shot Image Generation | AFHQ-Wild 10-shot | 34 | Feb 26, 2026 |
|---|