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 |
|---|---|---|---|---|---|
| Sentiment Transformation | IMDB NegToPos 1.0 (test) | 30 | Feb 26, 2026 | ||
| Sentiment Transformation | Sentiment Transformation Task | 30 | Feb 26, 2026 | ||
| Toxicity Mitigation | Toxicity Mitigation Task | 30 | Feb 26, 2026 | ||
| Toxicity mitigation | ToxicTop | 30 | Feb 26, 2026 | ||
| Toxicity mitigation | ToxicRandom | 30 | Feb 26, 2026 | ||
| Long Sequence Forecasting |
| Electricity |
| 30 |
| Jun 11, 2026 |
| Text-to-motion retrieval | HumanML3D (test) | 30 | Mar 26, 2026 |
|---|
| 3D Visual Question Answering | SQA3D | 30 | Jun 9, 2026 |
|---|
| Action Recognition | UAV-Human (CSv2) | 30 | Jun 23, 2026 |
|---|
| Image-to-text retrieval | SugarCrepe | 30 | May 26, 2026 |
|---|
| Game Playing | Go (round-robin) | 30 | Feb 26, 2026 |
|---|
| Vision-and-Language Navigation | REVERIE (test) | 30 | Mar 11, 2026 |
|---|
| Metric Depth Estimation | SUN RGB-D | 30 | Jun 1, 2026 |
|---|
| Trajectory Prediction | GCS | 30 | Feb 26, 2026 |
|---|
| Clustering | DIGIT | 30 | Jun 5, 2026 |
|---|
| Gait Recognition | CCPG (test) | 30 | Apr 30, 2026 |
|---|
| Visual Localization | 7scenes indoor | 30 | Feb 27, 2026 |
|---|
| Watermark Attack Robustness | Vicuna 7b 16k v1.5 (test) | 30 | Feb 26, 2026 |
|---|
| Image Denoising | BSD68 | 30 | Mar 17, 2026 |
|---|
| Image-Text Retrieval | MSCOCO 5K | 30 | May 7, 2026 |
|---|
| Camera Relocalization | 7-Scenes (test) | 30 | Feb 26, 2026 |
|---|
| Image Classification | ImageNet-C Severity Level 5, Mixed Shifts v1 (test) | 30 | Apr 21, 2026 |
|---|
| Text-to-Video Retrieval | MSRVTT (MSR) zero-shot | 30 | Apr 6, 2026 |
|---|
| Short-term time series forecasting | M4 weighted average from all datasets (test) | 30 | Feb 26, 2026 |
|---|
| Interpolation | PhysioNet 6 min quantization 2012 (test) | 30 | Feb 26, 2026 |
|---|