Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 2165 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Semantic Segmentation | FBP | 16 | May 15, 2026 | ||
| Semantic Segmentation | Potsdam | 16 | May 15, 2026 | ||
| Few-Shot Segmentation | iSAID (Target), PASCAL VOC (Source) | 16 | May 15, 2026 | ||
| Few-Shot Segmentation | LoveDA (Target), PASCAL VOC (Source) | 16 | May 15, 2026 | ||
| Few-Shot Segmentation | FBP (Target) PASCAL VOC (Source) | 16 | May 15, 2026 | ||
| Few-Shot Segmentation | Potsdam Target PASCAL VOC Source | 16 | May 15, 2026 | ||
| End-to-End Autonomous Driving | Bench2Drive base V0.0.3 (train) | 16 | May 15, 2026 | ||
| Body System Classification | Clinical Notes | 16 | May 15, 2026 | ||
| ROS Entity Recognition | Clinical Notes | 16 | May 15, 2026 | ||
| Motion Deblurring | FFHQ (val) | 16 | May 27, 2026 | ||
| Continual Visual Instruction Tuning | CVIT Single-type instruction | 16 | May 15, 2026 | ||
| Single Image Rain Removal | Rain100L 51 (test) | 16 | May 14, 2026 | ||
| VMAT Plan Optimization | REQUITE v1 (test) | 16 | May 14, 2026 | ||
| Discontinuous constituent parsing | NEGRA (test) | 16 | May 14, 2026 | ||
| Discontinuous constituent parsing | TIGER (test) | 16 | May 14, 2026 | ||
| Image Classification | Knee (test) | 16 | May 14, 2026 | ||
| Hallucination Evaluation | POPE adversarial (400-sample held-out set) | 16 | May 14, 2026 | ||
| Multi-Reward Alignment | Parti-Prompts | 16 | May 14, 2026 | ||
| Shape Registration | TOPKIDS (test) | 16 | May 14, 2026 | ||
| Skill Knowledge | Lens | 16 | May 14, 2026 | ||
| Scientific Reasoning | Lens | 16 | May 14, 2026 | ||
| Optical Character Recognition | Lens | 16 | May 14, 2026 | ||
| Relation Extraction | Lens | 16 | May 14, 2026 | ||
| Object Existence | Lens | 16 | May 14, 2026 | ||
| Object Counting | Lens | 16 | May 14, 2026 |