Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 566 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Low-shot Image Classification | ImageNet 1k (val) | 40 | Feb 26, 2026 | ||
| Link Prediction | WordNet noun hierarchy (transitive closure) (test) | 40 | Feb 26, 2026 | ||
| Text Classification | DBpedia (test) | 40 | Jun 2, 2026 | ||
| Camera Localization | 7-Scenes Chess | 40 | Feb 26, 2026 | ||
| Video Object Segmentation | SegTrack v2 (test) | 40 | Feb 26, 2026 | ||
| 3D Human Pose and Mesh Recovery | Human3.6M | 40 | Feb 26, 2026 | ||
| Image Classification | MNIST | 40 | Feb 26, 2026 | ||
| Face Identification | MegaFace 1M distractors 1.0 (test) | 40 | Feb 26, 2026 | ||
| Semantic Role Labeling | CoNLL Brown 2005 (test) | 40 | Feb 26, 2026 | ||
| Saliency Detection | DUT-OMRON | 40 | Feb 26, 2026 | ||
| Single Image Super-Resolution | Urban100 scale 2 | 40 | Feb 26, 2026 | ||
| Image Classification | CIFAR-100 (test) | 40 | Feb 26, 2026 | ||
| Question Answering | NewsQA (test) | 40 | Jun 12, 2026 | ||
| Object Counting | PASCAL VOC Count 2007 (test) | 40 | Feb 26, 2026 | ||
| Image Quality Assessment | TID 2008 | 40 | May 1, 2026 | ||
| Sentence Classification | Stanford Sentiment Treebank (SST) fine-grained (test) | 40 | Feb 26, 2026 | ||
| 3D Object Detection (Cars) | KITTI (test) | 40 | Feb 26, 2026 | ||
| Sentiment Analysis | Yelp P. (test) | 40 | Feb 26, 2026 | ||
| Molecule Property Prediction | HIV MoleculeNet (test) | 40 | Apr 23, 2026 | ||
| Nuclei Detection | PanNuke averaged across three dataset splits | 40 | Mar 31, 2026 | ||
| Accuracy of components | Hierarchical simulated dataset (test) | 40 | Feb 26, 2026 | ||
| Reproducibility | Real fMRI dataset (split-sample) | 40 | Feb 26, 2026 | ||
| Regression | Concrete UCI (test) | 40 | Jun 25, 2026 | ||
| Image Classification | Tiny-ImageNet Dirichlet alpha=0.1 (test) | 40 | Jul 7, 2026 | ||
| Depth Estimation | Cityscapes (test) | 40 | May 18, 2026 |