Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 1625 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Kernel State Reconstruction | tumor Student-t noise (test) | 20 | Feb 26, 2026 | ||
| Kernel State Reconstruction | tmdd lite Student-t noise (test) | 20 | Feb 26, 2026 | ||
| Kernel State Reconstruction | glucose Student-t noise (test) | 20 | Feb 26, 2026 | ||
| Kernel State Reconstruction | diabetes Student-t noise (test) | 20 | Feb 26, 2026 | ||
| State Reconstruction | 12 different tasks Student-t heavy-tailed noise | 20 | Feb 26, 2026 | ||
| State Reconstruction | 12 different tasks Correlated Gaussian noise | 20 | Feb 26, 2026 | ||
| State Reconstruction | 12 different tasks Isotropic Gaussian noise | 20 | Feb 26, 2026 | ||
| Autonomous Driving Planning | NAVSIM Navhard v2 (Stage 1) | 20 | May 19, 2026 | ||
| Adversarial Attack | Lichess-16M | 20 | Feb 26, 2026 | ||
| Adversarial Attack | Stockfish-8M | 20 | Feb 26, 2026 | ||
| Adversarial Attack | Millionbase 500k | 20 | Feb 26, 2026 | ||
| Adversarial Attack | Random-10M | 20 | Feb 26, 2026 | ||
| Adversarial Attack | Random-2M | 20 | Feb 26, 2026 | ||
| Adversarial Attack | Random-500k | 20 | Feb 26, 2026 | ||
| Federated Unlearning | Fashion-MNIST | 20 | Feb 26, 2026 | ||
| Federated Unlearning | CIFAR-10 | 20 | Feb 26, 2026 | ||
| Federated Unlearning | MNIST (test) | 20 | Feb 26, 2026 | ||
| Federated Unlearning | Fashion-MNIST (test) | 20 | Feb 26, 2026 | ||
| Image Classification | CIFAR-10 (test) | 20 | Feb 26, 2026 | ||
| Machine Unlearning | CIFAR-100 (test) | 20 | Feb 26, 2026 | ||
| Counterfactual Explanation Generation | Parkinsons | 20 | Feb 26, 2026 | ||
| SUM-M | CIFAR-10 n=10K | 20 | Feb 26, 2026 | ||
| SUM-M | MNIST n=100 | 20 | Feb 26, 2026 | ||
| Image Classification | MNIST PGL(3, R) (test) | 20 | Feb 26, 2026 | ||
| Robustness against harmful content generation | LMSYS harmful queries | 20 | Feb 26, 2026 |