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 |
|---|---|---|---|---|---|
| Audio-visual speech separation | LRS2-2Mix (test) | 33 | Feb 26, 2026 | ||
| Class-Incremental Learning | Stanford Cars CIL, T=10 (test) | 33 | May 12, 2026 | ||
| Feature Matching | WxBS | 33 | Jul 7, 2026 | ||
| Graph Classification | IMDB-B | 33 | Apr 1, 2026 | ||
| General Reasoning | AGIEval | 33 | Feb 26, 2026 | ||
| Spatio-temporal forecasting |
|---|
| NYC Citi Bike (test) |
| 33 |
| Jun 5, 2026 |
| Aggregate Tabular Benchmarking | Aggregate | 33 | Feb 26, 2026 |
|---|
| Anomaly Detection | ADGym Average performance across 14 domains | 33 | Feb 26, 2026 |
|---|
| Anomaly Detection | Real-world datasets | 33 | Feb 26, 2026 |
|---|
| Polyp Segmentation | EndoScene (test) | 33 | Feb 26, 2026 |
|---|
| Classification | VinDR-CXR | 33 | Jun 4, 2026 |
|---|
| Event Prediction | Chicago Crime | 33 | Feb 26, 2026 |
|---|
| Event Prediction | Chicago Accident | 33 | Feb 26, 2026 |
|---|
| 8-class classification | BloodMNIST | 33 | Mar 24, 2026 |
|---|
| Medical Image Classification | RetinaMNIST v2 (test) | 33 | Feb 26, 2026 |
|---|
| 3D Part Segmentation | ShapeNetPart (val) | 33 | Mar 17, 2026 |
|---|
| Referring Video Object Segmentation | Ref-Youtube-VOS v1.0 (test) | 33 | Feb 26, 2026 |
|---|
| Semantic Segmentation | Cityscapes | 33 | May 6, 2026 |
|---|
| Depth Estimation | DIODE (test) | 33 | Feb 26, 2026 |
|---|
| Segmentation | CAD-CAP WCE (5-fold cross-validation) | 33 | Feb 26, 2026 |
|---|
| Automatic Diagnosis | Muzhi | 33 | Feb 26, 2026 |
|---|
| 4x Super-resolution | FFHQ 256x256 (val) | 33 | Jun 30, 2026 |
|---|
| 3D Triangulation | ETH3D (train) | 33 | Mar 31, 2026 |
|---|
| Medical Image Segmentation | LiTS | 33 | Apr 15, 2026 |
|---|
| Instance Segmentation | LVIS 1.0 (val) | 33 | Apr 13, 2026 |
|---|