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 |
|---|---|---|---|---|---|
| Controllability | LLM Judge | 40 | Feb 26, 2026 | ||
| Controllability | PolyGuard | 40 | Feb 26, 2026 | ||
| Controllability | HarmBench | 40 | Feb 26, 2026 | ||
| Classification | Waterbirds (test) | 40 | Jun 24, 2026 | ||
| Image Classification | Masks | 40 | Feb 26, 2026 | ||
| Mathematical Reasoning |
|---|
| MATH500 |
| 40 |
| Apr 29, 2026 |
| Anomaly Detection | Lympho | 40 | Apr 9, 2026 |
|---|
| Anomaly Detection | Amazon | 40 | Jun 18, 2026 |
|---|
| Zero-Shot Posterior Sampling | IMAGENET 256x256 (val) | 40 | Feb 26, 2026 |
|---|
| Zero-Shot Posterior Sampling | FFHQ 256x256 (val) | 40 | Feb 26, 2026 |
|---|
| Time Series Forecasting | Electricity | 40 | May 21, 2026 |
|---|
| Reward Modeling | Unified Feedback (UF) | 40 | Feb 26, 2026 |
|---|
| Recommendation | Clothing | 40 | Apr 7, 2026 |
|---|
| Multilingual Retrieval | MTEB Multilingual v2 | 40 | May 14, 2026 |
|---|
| Information Retrieval | Natural Questions | 40 | Jul 3, 2026 |
|---|
| Few-shot 3D point cloud semantic segmentation | S3DIS (Mean across folds) | 40 | Mar 23, 2026 |
|---|
| Semantic Segmentation | Cityscapes-C long-term adaptation (target) | 40 | Feb 26, 2026 |
|---|
| Camera Pose Estimation | ScanNet static indoor scenes | 40 | May 18, 2026 |
|---|
| Text Visual Question Answering (Open Ended) | RIO-Bench Text-VQA (OE) Attack | 40 | Feb 26, 2026 |
|---|
| Text Visual Question Answering (Open Ended) | RIO-Bench Text-VQA (OE) Clean | 40 | Feb 26, 2026 |
|---|
| Object Visual Question Answering (Multiple Choice) | RIO-Bench Obj-VQA (MC) (Attack) | 40 | Feb 26, 2026 |
|---|
| Object Visual Question Answering (Multiple Choice) | RIO-Bench Obj-VQA (MC) Clean | 40 | Feb 26, 2026 |
|---|
| Image Classification | CIFAR-10 (test) | 40 | Feb 26, 2026 |
|---|
| Image Classification | CIFAR-10 (test) | 40 | Feb 26, 2026 |
|---|
| Clustering | Caltech 7-5V | 40 | Feb 26, 2026 |
|---|