Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 1960 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Topic Classification | NYT (test) | 18 | Feb 26, 2026 | ||
| Image Matting | P3M-500-P normal images | 18 | Feb 26, 2026 | ||
| Object Detection | LVIS (minival5k) | 18 | Feb 26, 2026 | ||
| Adversarial Attack | ED | 18 | Feb 26, 2026 | ||
| Adversarial Attack | CV2 | 18 | Feb 26, 2026 | ||
| Adversarial Attack | PC | 18 | Feb 26, 2026 | ||
| Adversarial Attack | BST | 18 | Feb 26, 2026 | ||
| Natural Language Understanding | GLUE official (val test) | 18 | Feb 26, 2026 | ||
| Class-Incremental Learning | ImageNet-100 10-task offline CIL setup (test) | 18 | Feb 26, 2026 | ||
| Oriented Object Detection | DOTA v2.0 (test-dev) | 18 | Feb 26, 2026 | ||
| Semantic Segmentation | Cityscapes (val) | 18 | Feb 26, 2026 | ||
| Depth Estimation | Structure3D (test) | 18 | Feb 26, 2026 | ||
| Audio-to-Visual Retrieval | MSR-VTT (test) | 18 | Feb 26, 2026 | ||
| Attribute Recognition | RAP zero-shot | 18 | Apr 30, 2026 | ||
| Visual Localization | Cambridge Landmarks Court | 18 | Apr 8, 2026 | ||
| Novel Pose Synthesis | InterHand Single hand → Single hand 2.6M (test) | 18 | Feb 26, 2026 | ||
| Virtual Try-on | VITON-HD high-resolution (test) | 18 | Apr 23, 2026 | ||
| Co-saliency Detection | CoSal 2015 (test) | 18 | Feb 26, 2026 | ||
| 3D Object Detection | OMNI3D OUT | 18 | Feb 26, 2026 | ||
| Zero-Shot Sketch-Based Image Retrieval | TU-Berlin | 18 | Feb 26, 2026 | ||
| Automatic Speech Recognition | LRS3 low-resource (test) | 18 | Feb 26, 2026 | ||
| 3D Object Detection | KITTI3D moderate difficulty (val) | 18 | Feb 26, 2026 | ||
| Object Detection | PASCAL VOC 2007 (test) | 18 | Feb 26, 2026 | ||
| Semantic Segmentation | PASCAL VOC20 without background class | 18 | Feb 26, 2026 | ||
| Super-resolution | DIV2K 46 (val) | 18 | Feb 26, 2026 |