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 |
|---|---|---|---|---|---|
| Mathematical Reasoning | SVAMP out-of-domain (test) | 50 | Feb 26, 2026 | ||
| Video-based Question Answering | STAR | 50 | Mar 20, 2026 | ||
| Image Classification | FEMNIST | 50 | Jul 1, 2026 | ||
| Red-Teaming (Attack Success Rate) | DANGEROUSQA | 50 | Mar 10, 2026 | ||
| Language Modeling | GovReport | 50 | Feb 26, 2026 | ||
| Image Classification |
| MNIST (test) |
| 50 |
| Feb 26, 2026 |
| Classification | MNIST | 50 | Feb 26, 2026 |
|---|
| Machine Translation | WMT En-De 2017 (test) | 50 | May 12, 2026 |
|---|
| Low-Light Image Enhancement | NPE | 50 | Apr 8, 2026 |
|---|
| Low-Light Image Enhancement | LIME | 50 | Apr 8, 2026 |
|---|
| Named Entity Recognition | multiNERD | 50 | Apr 21, 2026 |
|---|
| Document Retrieval | MS MARCO | 50 | Jul 3, 2026 |
|---|
| Hyperspectral Image Denoising | ICVL (test) | 50 | Feb 26, 2026 |
|---|
| Salient Object Detection | VT5000 | 50 | Apr 13, 2026 |
|---|
| 6D Pose and Size Estimation | REAL275 | 50 | Feb 26, 2026 |
|---|
| Panoptic Segmentation | ADE20K | 50 | May 20, 2026 |
|---|
| 1-way Few-shot Semantic Segmentation | PASCAL-5i 1-way | 50 | Feb 26, 2026 |
|---|
| Link Prediction | Computers | 50 | Jun 11, 2026 |
|---|
| Backdoor Attack Classification | HSOL | 50 | Mar 18, 2026 |
|---|
| Sentiment Analysis | SST-2 (test) | 50 | Feb 26, 2026 |
|---|
| Reinforcement Learning | Atari 100k | 50 | Jul 7, 2026 |
|---|
| Image Classification | CIFAR100 | 50 | Mar 20, 2026 |
|---|
| Human Mesh Reconstruction | Human3.6M | 50 | Feb 26, 2026 |
|---|
| Molecular property prediction | SIDER | 50 | Jun 5, 2026 |
|---|
| Image Classification | Clothing1M | 50 | Jun 11, 2026 |
|---|