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 |
|---|---|---|---|---|---|
| Biophysical Parameter Estimation | Morris-Lecar biophysical model Hopf regime, 1% noise | 56 | Mar 11, 2026 | ||
| Parameter Estimation | Morris-Lecar model Homoclinic regime (1% noise) | 56 | Mar 11, 2026 | ||
| Jailbreak Attack Defense | MM-SafetyBench | 56 | Apr 2, 2026 | ||
| Regression | cpu_act | 56 | Jun 12, 2026 | ||
| Video Understanding | Video-MME v1.0 (test) |
| 56 |
| Mar 6, 2026 |
| Image Classification | CIFAR-10 | 56 | Mar 5, 2026 |
|---|
| Out-of-Distribution Detection | Places OOD ImageNet-1k ID | 56 | Jun 17, 2026 |
|---|
| Visualization Generation | MatPlotAgent | 56 | Mar 4, 2026 |
|---|
| Mathematical Reasoning | Math500 | 56 | Jun 9, 2026 |
|---|
| Detection Efficiency | Alpaca OnlyTarget Long (malicious) | 56 | Mar 4, 2026 |
|---|
| Detection Efficiency | Alpaca OnlyTarget Long (benign) | 56 | Mar 4, 2026 |
|---|
| Targeted Attack Detection | Alpaca OnlyTarget Short | 56 | Mar 4, 2026 |
|---|
| Targeted Attack Detection | Alpaca OnlyTarget Medium | 56 | Mar 4, 2026 |
|---|
| Multimodal Mathematical Reasoning | OlympiadBench | 56 | Apr 10, 2026 |
|---|
| Multispectral Image Denoising | Toys & Face MSI 256 x 256 x 31 (test) | 56 | Mar 4, 2026 |
|---|
| Breast density prediction | EMBED | 56 | Mar 4, 2026 |
|---|
| Privacy Editing | TDE Email | 56 | Feb 27, 2026 |
|---|
| Molecular property prediction | HIV (scaffold split) | 56 | Jun 25, 2026 |
|---|
| Molecular property prediction | Tox21 scaffold split | 56 | Jun 25, 2026 |
|---|
| Video Understanding | TGIF | 56 | Jun 24, 2026 |
|---|
| Image Understanding | MMStar | 56 | Jun 4, 2026 |
|---|
| Action Anticipation | NTURGBD | 56 | Feb 26, 2026 |
|---|
| Action Anticipation | UTKinects | 56 | Feb 26, 2026 |
|---|
| Action Anticipation | DARai Fine-grained | 56 | Feb 26, 2026 |
|---|
| Physical Commonsense Reasoning | PIQA | 56 | Mar 4, 2026 |
|---|