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 |
|---|---|---|---|---|---|
| Deep Research | BrowseComp-ZH (BC-zh) original (test) | 45 | Feb 26, 2026 | ||
| Grid-world Navigation | FrozenLake reward reversal Hidden drift | 45 | Feb 26, 2026 | ||
| Knowledge Evaluation | Natural Questions (NQ) (Evaluation) | 45 | Apr 6, 2026 | ||
| Jailbreak Attack | Prefilling Attack 40 tokens | 45 | Feb 26, 2026 | ||
| Jailbreak Attack | Prefilling Attack 20 tokens |
| 45 |
| Feb 26, 2026 |
| Jailbreak Attack | Prefilling Attack 10 tokens | 45 | Feb 26, 2026 |
|---|
| Reasoning | Reasoning Suite Average | 45 | Apr 28, 2026 |
|---|
| Text Classification | BANKING77 Dir(0.01) (test) | 45 | Feb 26, 2026 |
|---|
| Spatio-Temporal Forecasting | TaxiBJ | 45 | Mar 25, 2026 |
|---|
| Anomaly Detection | Br35H | 45 | Apr 14, 2026 |
|---|
| Trajectory Prediction | ETH UCY (Univ) | 45 | Jun 17, 2026 |
|---|
| Video Reconstruction | WebVid 10M | 45 | Apr 21, 2026 |
|---|
| End-to-end planning | nuScenes | 45 | May 13, 2026 |
|---|
| Visual Odometry | KITTI | 45 | Apr 27, 2026 |
|---|
| Mesh Saliency Prediction | Proposed Dataset Non-textured 1.0 (test) | 45 | Feb 26, 2026 |
|---|
| Mesh Saliency Prediction | Proposed Dataset Textured Mesh 1.0 (test) | 45 | Feb 26, 2026 |
|---|
| Medical Image Segmentation | COVID-CT | 45 | Mar 19, 2026 |
|---|
| Image Fitting | Kodak (test) | 45 | Apr 29, 2026 |
|---|
| Video Question Answering | VideoMME | 45 | Apr 21, 2026 |
|---|
| Few-shot classification | CPTAC k=25 positive samples (test) | 45 | Feb 26, 2026 |
|---|
| Few-shot Classification | CPTAC | 45 | Feb 26, 2026 |
|---|
| Vision Understanding | MMVP | 45 | Jun 23, 2026 |
|---|
| Style Transfer | MS-COCO (content) + WikiArt (style) (test) | 45 | May 27, 2026 |
|---|
| Knowledge Editing | VLKEB | 45 | Feb 26, 2026 |
|---|
| Image Classification | UCF | 45 | May 7, 2026 |
|---|