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 |
|---|---|---|---|---|---|
| Relative Robustness Analysis | Combined Past Tense, OR-Bench, MMLU | 36 | Apr 15, 2026 | ||
| Global Counterfactual Explanations | HELOC | 36 | Apr 15, 2026 | ||
| Global Counterfactual Explanations | German Credit | 36 | Apr 15, 2026 | ||
| Global Counterfactual Explanations | Default Credit | 36 | Apr 15, 2026 | ||
| Global Counterfactual Explanations | COMPAS |
| 36 |
| Apr 15, 2026 |
| Image Classification | Food-101 (test) | 36 | Jun 30, 2026 |
|---|
| Speech Recognition | Switchboard eval00 | 36 | Apr 14, 2026 |
|---|
| Face Verification | RFW | 36 | Jun 4, 2026 |
|---|
| Medical Diagnostic Reasoning | MedCaseReasoning official 897-case (test) | 36 | Apr 14, 2026 |
|---|
| Creative Writing | Creative Writing | 36 | Apr 14, 2026 |
|---|
| Question Answering | SQuAD KRE-curated version | 36 | Apr 14, 2026 |
|---|
| Open-ended Question Answering | ConFiQA (test) | 36 | Apr 14, 2026 |
|---|
| Multi-step Reasoning Question Answering | ConFiQA MR (test) | 36 | Apr 14, 2026 |
|---|
| Poisoning | CIFAR100 | 36 | Apr 14, 2026 |
|---|
| Poisoning | CIFAR10 | 36 | Apr 14, 2026 |
|---|
| Artistic Style Erasure | SD Other Class artistic styles 1.4 (test) | 36 | Apr 14, 2026 |
|---|
| Artistic Style Erasure | SD Target Class artistic styles 1.4 (test) | 36 | Apr 14, 2026 |
|---|
| Linear Regression | Equi-correlation Sigma_a r=0.3 | 36 | Apr 14, 2026 |
|---|
| Linear Regression | Equi-correlation Sigma_a r=0.1 | 36 | Apr 14, 2026 |
|---|
| Linear regression | Linear regression with Toeplitz Sigma_a r=0.6 | 36 | Apr 14, 2026 |
|---|
| Linear regression | Linear regression with Toeplitz Sigma_a r=0.4 | 36 | Apr 14, 2026 |
|---|
| Partial Multi-Label Learning | reference | 36 | Apr 13, 2026 |
|---|
| Partial Multi-Label Learning | arts | 36 | Apr 13, 2026 |
|---|
| Partial Multi-Label Learning | reference | 36 | Apr 13, 2026 |
|---|
| Partial Multi-Label Learning | arts | 36 | Apr 13, 2026 |
|---|