Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 1032 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Uncertainty Calibration | CIFAR-10.1 C | 27 | Feb 26, 2026 | ||
| Uncertainty Calibration | CIFAR-10.1 | 27 | Feb 26, 2026 | ||
| Image Harmonization | iHarmony4 | 27 | Feb 26, 2026 | ||
| Misclassification Detection | CIFAR-100 | 27 | May 26, 2026 | ||
| Homography Estimation | HPatches (Viewpoint) | 27 | Feb 26, 2026 | ||
| Semantic Segmentation | PASCAL Context without background class 59 | 27 | Feb 26, 2026 | ||
| Visual Question Answering | Slake | 27 | Feb 26, 2026 | ||
| Scene Flow Estimation | VoD (View-of-Delft) (test) | 27 | Apr 29, 2026 | ||
| Novel Class Discovery | CIFAR-100 | 27 | Jul 7, 2026 | ||
| Conditional layout generation (Category to Size and Position) | PubLayNet | 27 | Feb 26, 2026 | ||
| Conditional layout generation (Category to Size and Position) | Rico | 27 | Feb 26, 2026 | ||
| Camera pose registration | NeRF Synthetic v1 (test) | 27 | Feb 26, 2026 | ||
| Domain Generalization | PACS, VLCS, OfficeHome, TerraIncognita, DomainNet | 27 | Feb 26, 2026 | ||
| Video Recognition | Something-Something V1 | 27 | Feb 26, 2026 | ||
| Atomization energy prediction | QM7 (10-fold cross validation) | 27 | Mar 25, 2026 | ||
| Image Classification | CIFAR-10 40% asymmetric noise | 27 | Feb 26, 2026 | ||
| Instruction Following | Super-Natural Instructions (test) | 27 | Mar 5, 2026 | ||
| Video Question Answering | TGIF-Frame (test) | 27 | Feb 27, 2026 | ||
| Image question answering | OCR-VQA | 27 | Feb 26, 2026 | ||
| Node Classification | Squirrel (10 different splits) | 27 | Feb 26, 2026 | ||
| Node Classification | Cornell (10 different splits) | 27 | Feb 26, 2026 | ||
| Node Classification | Texas (10 different splits) | 27 | Feb 26, 2026 | ||
| Node Classification | Citeseer (10 different splits) | 27 | Feb 26, 2026 | ||
| Traveling Salesman Problem | TSP N=200 | 27 | May 26, 2026 | ||
| Image Enhancement | Adobe Five-K | 27 | Apr 10, 2026 |