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 Classification | MNIST Decentralized, 2 m.n. (test) | 84 | Jun 26, 2026 | ||
| Image Classification | MNIST Decentralized 1 m.n. (test) | 84 | Jun 26, 2026 | ||
| Image Classification | MNIST Decentralized 0 m.n. (test) | 84 | Jun 26, 2026 | ||
| Image Classification | MNIST Centralized (test) | 84 | Jun 26, 2026 | ||
| Predict-then-Optimize | 2D knapsack (test) | 84 | Jun 19, 2026 |
| Multi-label Classification | Multi-label datasets | 84 | Jun 5, 2026 |
|---|
| Multi-phase Image Segmentation | Image scenario | 84 | Jun 5, 2026 |
|---|
| Instructional Image Editing | CV-Arena 12K examples 1.0 | 84 | Jun 2, 2026 |
|---|
| Object Detection | HRIPCB | 84 | Jun 2, 2026 |
|---|
| Multimodal Understanding | MMMU, SEED, OCRBench, VizWiz, ScienceQA, and TextVQA Average (test) | 84 | May 26, 2026 |
|---|
| Data Ranking | Mistake math | 84 | May 21, 2026 |
|---|
| Long-term Conversation Question Answering | Realtalk | 84 | May 21, 2026 |
|---|
| Long-context Question Answering | LongMemEval LongConvQA | 84 | May 21, 2026 |
|---|
| Semantic Hijacking Attack Success Evaluation | Semantic Hijacking Evaluation Set | 84 | May 19, 2026 |
|---|
| Federated Learning | Roman-empire | 84 | May 8, 2026 |
|---|
| Federated Learning | Cora | 84 | May 8, 2026 |
|---|
| Safety Evaluation | XSTest Unsafe | 84 | Jun 26, 2026 |
|---|
| Fixed-time learning | Heat Equation 24 (test) | 84 | Apr 28, 2026 |
|---|
| PDE Learning | Non-linear Fokker-Planck equation (test) | 84 | Apr 28, 2026 |
|---|
| Safety Evaluation | DirectHarm | 84 | Apr 24, 2026 |
|---|
| Blind Image Deblurring | Levin dataset | 84 | Apr 23, 2026 |
|---|
| Image Manipulation Localization | CAT-Net evaluation protocol (test) | 84 | Apr 13, 2026 |
|---|
| NACE industry classification | MONETA | 84 | Apr 10, 2026 |
|---|
| Speech Translation Metric Evaluation | Europarl-ST (test) | 84 | Apr 9, 2026 |
|---|
| Code Generation | LiveCodeBench | 84 | May 29, 2026 |
|---|