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 |
|---|---|---|---|---|---|
| AI-generated image detection | SD v1.5 | 36 | Mar 4, 2026 | ||
| Multivariate Time Series Forecasting | METR-LA (test) | 36 | Feb 27, 2026 | ||
| Accuracy Prediction | NAS-Bench-201 8 (whole dataset) | 36 | Feb 27, 2026 | ||
| Web Navigation | WebShop (test) | 36 | Jun 11, 2026 | ||
| Locomotion | D4RL Ant medium-offline | 36 | Feb 26, 2026 |
| Locomotion | D4RL Hopper medium-offline | 36 | Feb 26, 2026 |
|---|
| Locomotion | D4RL Walker2d medium-offline | 36 | Feb 26, 2026 |
|---|
| Locomotion | D4RL HalfCheetah medium-offline | 36 | Feb 26, 2026 |
|---|
| Self-evaluation | CVBench | 36 | Feb 26, 2026 |
|---|
| Self-evaluation | VisualCoT | 36 | Feb 26, 2026 |
|---|
| Self-evaluation | MMVet | 36 | Feb 26, 2026 |
|---|
| Self-evaluation | ViLP | 36 | Feb 26, 2026 |
|---|
| Weakly Supervised Classification | TCGA-BRCA Few-shot | 36 | Feb 26, 2026 |
|---|
| Weakly Supervised Classification | TCGA-NSCLC Few-shot | 36 | Feb 26, 2026 |
|---|
| Weakly Supervised Classification | CAMELYON Few-shot | 36 | Feb 26, 2026 |
|---|
| Manipulation Localization | FFHQ OOD 1.0 | 36 | Feb 26, 2026 |
|---|
| Open-World Object Detection | M-OWODB Task 1 | 36 | Jul 7, 2026 |
|---|
| Image Generation | ImageNet-256 (FID-50K) | 36 | May 20, 2026 |
|---|
| Reasoning | MMLU (test) | 36 | May 19, 2026 |
|---|
| 3DGS Compression | Tank&Temples | 36 | May 21, 2026 |
|---|
| Anomaly Detection | MPDD | 36 | Apr 6, 2026 |
|---|
| Multimodal Classification | KS | 36 | Jul 9, 2026 |
|---|
| Adversarial Attack | Breakout Black-box discrete (test) | 36 | Feb 26, 2026 |
|---|
| Adversarial Attack | Breakout White-box discrete (test) | 36 | Feb 26, 2026 |
|---|
| Node Classification | Twitch (OOD) | 36 | Feb 26, 2026 |
|---|