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 |
|---|---|---|---|---|---|
| Image Retrieval | P → R (test) | 52 | Feb 26, 2026 | ||
| Image Retrieval | A → D (test) | 52 | Feb 26, 2026 | ||
| Image Retrieval | MNIST → USPS (test) | 52 | Feb 26, 2026 | ||
| Cross-domain image retrieval | Office-31 A → W | 52 | Feb 26, 2026 | ||
| Cross-domain image retrieval | Office-31 A → D |
| 52 |
| Feb 26, 2026 |
| Cross-domain image retrieval | MNIST → USPS | 52 | Feb 26, 2026 |
|---|
| Multi-class classification | CIFAR-100 (test) | 52 | May 25, 2026 |
|---|
| Time Series Forecasting | ETTh1 | 52 | Feb 26, 2026 |
|---|
| Zero-shot Question Answering and Reasoning | Accuracy Tasks Zero-shot (AC, AE, WI, QA) | 52 | Feb 26, 2026 |
|---|
| Spatiotemporal Traffic Forecasting | Orange (test) | 52 | Feb 26, 2026 |
|---|
| Spatiotemporal Traffic Forecasting | Contra Costa (test) | 52 | Feb 26, 2026 |
|---|
| Spatiotemporal Traffic Forecasting | Alameda (test) | 52 | Feb 26, 2026 |
|---|
| Multi-answer Question Answering | MAQA | 52 | May 12, 2026 |
|---|
| Safety Alignment | Do-Not-Answer | 52 | Mar 10, 2026 |
|---|
| Tabular Anomaly Detection | Vertebral | 52 | Jun 4, 2026 |
|---|
| Question Answering | MMLU | 52 | Feb 26, 2026 |
|---|
| Graph Classification | DrugOOD EC50 (OOD test) | 52 | Feb 26, 2026 |
|---|
| Time Series Forecasting | Solar-Energy | 52 | May 13, 2026 |
|---|
| Time Series Forecasting | ETT | 52 | Mar 11, 2026 |
|---|
| General Question Answering | NQ | 52 | Apr 16, 2026 |
|---|
| NC vs. AD classification | ADNI | 52 | Jun 19, 2026 |
|---|
| Autonomous Driving Planning | NAVSIM v2 (test) | 52 | Jun 1, 2026 |
|---|
| Chart-based Question Answering | ChartQA Pro | 52 | Apr 6, 2026 |
|---|
| Image Classification | CIFAR10-C | 52 | Jul 1, 2026 |
|---|
| Image Classification | Country211 | 52 | Jun 8, 2026 |
|---|