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 |
|---|---|---|---|---|---|
| 5-way classification | ChestX (test) | 30 | Mar 6, 2026 | ||
| Low-light image enhancement | SDE in | 30 | Jun 30, 2026 | ||
| Untargeted Adversarial Attack | Flickr30K 1,000 images (test) | 30 | Mar 6, 2026 | ||
| Untargeted Adversarial Attack | Flickr30K | 30 | Mar 6, 2026 | ||
| Targeted Adversarial Attack | ImageNet |
| 30 |
| Mar 6, 2026 |
| Pre-training data detection | BookMIA (test) | 30 | Mar 6, 2026 |
|---|
| Pre-training data detection | BookTection (test) | 30 | Mar 6, 2026 |
|---|
| Pre-training data detection | WikiMIA (test) | 30 | Mar 6, 2026 |
|---|
| Pre-training data detection | MIMIR | 30 | Mar 6, 2026 |
|---|
| Multilingual Language Understanding | MMMLU | 30 | Mar 6, 2026 |
|---|
| Spatiotemporal Field Reconstruction | Navier-Stokes 10% Subset | 30 | Mar 6, 2026 |
|---|
| Spatiotemporal Field Reconstruction | Navier-Stokes (Full) | 30 | Mar 6, 2026 |
|---|
| RGB-Event Object Tracking | FE108 | 30 | Jun 26, 2026 |
|---|
| Refusal Ablation | LlamaGuard | 30 | Mar 5, 2026 |
|---|
| Homography Estimation | PDSCOCO | 30 | Mar 5, 2026 |
|---|
| Homography Estimation | GoogleEarth | 30 | Mar 5, 2026 |
|---|
| Math Reasoning | MathVista | 30 | May 19, 2026 |
|---|
| Repeater Prediction | Avs | 30 | Mar 5, 2026 |
|---|
| Churn Prediction | Amazon | 30 | Mar 5, 2026 |
|---|
| Hallucination Detection | XTRUST | 30 | Mar 5, 2026 |
|---|
| Shape Reconstruction | Novel Object Set Princeton and Stanford repositories | 30 | Mar 5, 2026 |
|---|
| Sentiment Analysis | Rotten Tomatoes (RT) | 30 | Mar 5, 2026 |
|---|
| Sentiment Analysis | SST2 | 30 | Mar 5, 2026 |
|---|
| Topic Classification | AGNews | 30 | Mar 5, 2026 |
|---|
| Question Classification | TREC | 30 | Mar 5, 2026 |
|---|