Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 1671 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Scene Text Image Super-Resolution | TextZoom hard | 20 | Feb 26, 2026 | ||
| Scene Text Image Super-Resolution | TextZoom (medium) | 20 | Feb 26, 2026 | ||
| Scene Text Image Super-Resolution | TextZoom easy | 20 | Feb 26, 2026 | ||
| Knowledge-grounded Dialog Generation | Wizard of Wikipedia (WoW) | 20 | Feb 26, 2026 | ||
| Knowledge-grounded Dialog Generation | MultiWOZ | 20 | Feb 26, 2026 | ||
| Universal Lesion Detection | DeepLesion official (test) | 20 | Feb 26, 2026 | ||
| Text Classification | OCNLI (dev) | 20 | Feb 26, 2026 | ||
| Text Classification | BQ Corpus (test) | 20 | Feb 26, 2026 | ||
| Text Classification | BQ Corpus (dev) | 20 | Feb 26, 2026 | ||
| Text Classification | XNLI (test) | 20 | Feb 26, 2026 | ||
| Anomaly Detection | ShanghaiTech | 20 | Feb 26, 2026 | ||
| Table Question Answering | WIKISQL WEAK (test) | 20 | Feb 26, 2026 | ||
| Deep Metric Learning | Stanford Online Products (SOP) | 20 | Feb 26, 2026 | ||
| Image Inpainting | Places2 512x512 (test) | 20 | Feb 26, 2026 | ||
| Knowledge Graph Completion | ogbl-wikikg2 (test) | 20 | Feb 26, 2026 | ||
| Facial Attribute Classification (Attractive) | CelebA (test) | 20 | Feb 26, 2026 | ||
| Pose Estimation | ShapeNet Cross-Category 1D | 20 | Feb 26, 2026 | ||
| Pose Estimation | ShapeNet Intra-Category 1D | 20 | Feb 26, 2026 | ||
| Question Answering | TyDiQA GoldP | 20 | Feb 26, 2026 | ||
| Video Frame Interpolation | Vimeo-90K septuplet | 20 | Feb 26, 2026 | ||
| Image Classification | CIFAR100 (IM=10) LT (test) | 20 | Feb 26, 2026 | ||
| Aspect-based Sentiment Analysis | Lap SemEval-2014 (test) | 20 | Apr 16, 2026 | ||
| Fine-grained Recognition | iNaturalist 2018 (test) | 20 | Feb 26, 2026 | ||
| Video Deblurring | DVD | 20 | Feb 26, 2026 | ||
| Head Pose Estimation | AFLW2000-3D | 20 | Feb 26, 2026 |