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 Reconstruction | MRI | 30 | May 26, 2026 | ||
| Uncertainty Estimation | Gaussian noise dataset | 30 | May 26, 2026 | ||
| Uncertainty Estimation | Poisson noise dataset | 30 | May 26, 2026 | ||
| Temporal Link Prediction | CollegeMsg (inductive) | 30 | May 26, 2026 | ||
| Temporal Link Prediction | UCI (inductive) | 30 | May 26, 2026 | ||
| Multi-turn Math | GSM8K multi-turn |
| 30 |
| May 26, 2026 |
| Multi-turn math reasoning | SHARDED | 30 | May 26, 2026 |
|---|
| CNOT circuit synthesis | 4 x 4 MDS matrices over GL(8, F2) | 30 | May 26, 2026 |
|---|
| Harmful Meme Detection | Ex-ToxiCN-MM (test) | 30 | May 26, 2026 |
|---|
| MDD Diagnosis | REST-meta-MDD site 20 (test) | 30 | May 26, 2026 |
|---|
| Vulnerability Detection | ZKAP Dataset | 30 | May 26, 2026 |
|---|
| Vulnerability Detection | Real-world ZK Circuits All | 30 | May 26, 2026 |
|---|
| Safety classification disagreement | WildChat Content | 30 | May 26, 2026 |
|---|
| Safety classification disagreement | WildChat Intent | 30 | May 26, 2026 |
|---|
| Multilingual Mathematical Reasoning | PolyMath (test) | 30 | May 22, 2026 |
|---|
| 10-step incompressible flow rollout | NS-SL extreme viscosity (x) (test) | 30 | May 22, 2026 |
|---|
| 10-step incompressible flow rollout | NS-SL moderate viscosity (m) (test) | 30 | May 22, 2026 |
|---|
| Machine Translation | WildMTBench | 30 | May 22, 2026 |
|---|
| Machine Translation | DomainMTBench | 30 | May 22, 2026 |
|---|
| Machine Translation | Mandarin ⇔ Minority (General) | 30 | May 22, 2026 |
|---|
| Machine Translation | FLORES-200 XX ⇔ XX | 30 | May 22, 2026 |
|---|
| Machine Translation | FLORES-200 ZH ⇔ XX | 30 | May 22, 2026 |
|---|
| Question Answering | NQ (test) | 30 | May 22, 2026 |
|---|
| RDP Auditing | CIFAR-10 | 30 | May 22, 2026 |
|---|
| RDP Auditing | MNIST | 30 | May 22, 2026 |
|---|