Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 2454 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Semantic Segmentation | SAR One-Domain-Gap (Unseen Polarization) | 15 | Mar 13, 2026 | ||
| Semantic Segmentation | SAR One-Domain-Gap (Unseen Region) | 15 | Mar 13, 2026 | ||
| Building Extraction | Unseen Region and Polarization (F2A) | 15 | Mar 13, 2026 | ||
| Video Depth Estimation | KITTI | 15 | Jun 26, 2026 | ||
| Video Depth Estimation | ScanNet | 15 | Jun 26, 2026 | ||
| Disentanglement | COIL 2 | 15 | Mar 13, 2026 | ||
| Disentanglement | COIL3 | 15 | Mar 13, 2026 | ||
| Disentanglement | FlatLand permutation colors | 15 | Mar 13, 2026 | ||
| Disentangled Representation Learning | FlatLand rotation colors | 15 | Mar 13, 2026 | ||
| Memorized Content Extraction | EchoTrace (arXiv) | 15 | Mar 13, 2026 | ||
| 3D Medical Report Generation | AMOS-MM 1.0 (test) | 15 | Mar 13, 2026 | ||
| Failure Detection | Close Box Combined (train test) | 15 | Mar 13, 2026 | ||
| Failure Detection | Close Box Out-of-Distribution (train test) | 15 | Mar 13, 2026 | ||
| Failure Detection | Close Box In-Distribution (train test) | 15 | Mar 13, 2026 | ||
| Clustering | Iris | 15 | May 12, 2026 | ||
| Class-Incremental Learning | TinyImageNet 10 Steps (test) | 15 | Mar 16, 2026 | ||
| Classification | German Credit | 15 | Apr 3, 2026 | ||
| VLA Training Throughput (pi_0) | ManiSkill | 15 | Mar 13, 2026 | ||
| VLA Training Throughput (GR00T N1.5) | LIBERO | 15 | Mar 13, 2026 | ||
| VLA Training Throughput (pi_0.5) | LIBERO | 15 | Mar 13, 2026 | ||
| Class-Incremental Learning | CIFAR100 | 15 | Mar 13, 2026 | ||
| Image Classification | MNIST to Fashion MNIST (test) | 15 | Mar 13, 2026 | ||
| Harmlessness Evaluation | HarmBench | 15 | Mar 12, 2026 | ||
| Safety Evaluation | BeaverTails (test) | 15 | Mar 12, 2026 | ||
| 3D Scene Reconstruction | RE10K (test) | 15 | Mar 27, 2026 |