Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 1664 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Semi-supervised node classification | ogbn-arxiv | 20 | Feb 26, 2026 | ||
| Masked Language Modeling | XSUM randomly sampled | 20 | Feb 26, 2026 | ||
| Masked Language Modeling | SNLI (randomly sampled) | 20 | Feb 26, 2026 | ||
| Automatic Speech Recognition | LibriSpeech 960h | 20 | Feb 26, 2026 | ||
| Few-shot Named Entity Recognition | FewNERD Inter 1.0 | 20 | Feb 26, 2026 | ||
| Long-tailed Image Classification | CIFAR-10-LT IF=100 (test) | 20 | Jun 2, 2026 | ||
| Scene Text Recognition | CUTE80 288 | 20 | Feb 26, 2026 | ||
| 3D Object Recognition | ShapeNet | 20 | Feb 26, 2026 | ||
| Generalized Object Detection | Pascal VOC 2007+2012 (Split-3) | 20 | Feb 26, 2026 | ||
| Generalized Object Detection | Pascal VOC 2007+2012 (Split-1) | 20 | Feb 26, 2026 | ||
| Few-Shot Open-Set Recognition | tieredImageNet (test) | 20 | Feb 26, 2026 | ||
| Single-label multi-class topic classification | SCOTUS (test) | 20 | Jun 30, 2026 | ||
| Image Classification | ImageNet 1k (val) | 20 | Feb 26, 2026 | ||
| political perspective detection | Allsides | 20 | Jun 8, 2026 | ||
| Multi-view Clustering | Cora | 20 | Feb 26, 2026 | ||
| Video Steganography | Vimeo-T200 (test) | 20 | Feb 26, 2026 | ||
| Grasping | Epic-Kitchens (Held-out Rare Objects) | 20 | Feb 26, 2026 | ||
| Node Classification | PubMed (test) | 20 | Feb 26, 2026 | ||
| Scene Graph Classification (SGCls) | Visual Genome (VG) | 20 | Feb 26, 2026 | ||
| Image Denoising | CBSD68 35 (test) | 20 | Feb 26, 2026 | ||
| Image Denoising | Kodak 28 (test) | 20 | Feb 26, 2026 | ||
| Image Super-resolution | CBSD68 | 20 | Feb 26, 2026 | ||
| Image Super-resolution | Kodak | 20 | Feb 26, 2026 | ||
| Video reconstruction | UVG (test) | 20 | Apr 9, 2026 | ||
| Optical Flow Estimation | KITTI (test) | 20 | Feb 26, 2026 |