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 |
|---|---|---|---|---|---|
| Part-of-Speech Tagging | POS (test) | 33 | Feb 26, 2026 | ||
| Image Classification | SVHN to MNIST standard (test) | 33 | Feb 26, 2026 | ||
| Trajectory Prediction | ETH v1.0 (test) | 33 | Feb 26, 2026 | ||
| Action Unit Detection | FERA17 (test) | 33 | Feb 26, 2026 | ||
| Action Unit Detection | FERA17 (val) | 33 | Feb 26, 2026 | ||
| Multivariate Time Series Classification |
| PenDigits |
| 33 |
| Feb 26, 2026 |
| Multivariate Time Series Classification | Libras | 33 | Feb 26, 2026 |
|---|
| Image Retrieval | Oxford Revisited (Hard) | 33 | Feb 26, 2026 |
|---|
| Summarization | CNN/DailyMail (test) | 33 | Mar 12, 2026 |
|---|
| Semantic Segmentation | Semantic3D reduced-8 (test) | 33 | Feb 27, 2026 |
|---|
| Node Classification | PubMed standard (fixed split) | 33 | Feb 26, 2026 |
|---|
| Image Classification | ILSVRC 2015 (val) | 33 | Feb 26, 2026 |
|---|
| Face Detection | WiderFace (test) | 33 | Jun 19, 2026 |
|---|
| Sentence Classification | CR (test) | 33 | Feb 26, 2026 |
|---|
| Open-domain question answering | Quasar-T (test) | 33 | Feb 26, 2026 |
|---|
| Person Re-identification | Market-1501 Single Query 1.0 | 33 | Feb 26, 2026 |
|---|
| Abnormal event detection | UCSD Ped1 (test) | 33 | Feb 26, 2026 |
|---|
| Zero-Shot Action Recognition | UCF101 (test) | 33 | Feb 26, 2026 |
|---|
| Classification | Caltech101 (test) | 33 | Feb 26, 2026 |
|---|
| Photo Aesthetics Classification | AVA (test) | 33 | Feb 26, 2026 |
|---|
| Sentiment Analysis | Yelp '13 (test) | 33 | Feb 26, 2026 |
|---|
| Image Question Answering | DAQUAR REDUCED (test) | 33 | Feb 26, 2026 |
|---|
| Phoneme Recognition | TIMIT (dev) | 33 | Jun 8, 2026 |
|---|
| Video Classification | Sports-1M | 33 | Feb 27, 2026 |
|---|
| Multi-target tracking | KITTI Pedestrian (test) | 33 | Feb 26, 2026 |
|---|