Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 1514 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Object Recognition | VisDA C | 21 | Feb 26, 2026 | ||
| Object Recognition | Office 59 (test) | 21 | Feb 26, 2026 | ||
| Multiple Object Tracking | Radiate good weather (train) | 21 | Feb 26, 2026 | ||
| Anomaly Detection | Optical | 21 | Feb 26, 2026 | ||
| Machine Reading Comprehension | DREAM (dev) | 21 | Feb 26, 2026 | ||
| Semantic Segmentation | Cityscapes 1/8 partition protocol (val) | 21 | Feb 26, 2026 | ||
| Multi-choice Question Answering | MMLU professional medicine | 21 | Feb 26, 2026 | ||
| Image Classification | Sun-297 (val) | 21 | Apr 3, 2026 | ||
| Object Detection | MS-COCO 48/17 base/novel | 21 | Feb 26, 2026 | ||
| Text-to-Video Retrieval | DiDeMo 1K videos (test) | 21 | Apr 14, 2026 | ||
| Link prediction | WikiKG90M v2 (test) | 21 | Jun 2, 2026 | ||
| Few-shot Image Classification | CifarFS 5-way 5-shot | 21 | Feb 26, 2026 | ||
| Pseudo Segmentation Label Generation | VOC 2012 (train) | 21 | Feb 26, 2026 | ||
| Image Inpainting | Places2 challenge dataset (test) | 21 | Feb 26, 2026 | ||
| Formality Transfer | Hindi Formality Transfer (val) | 21 | Feb 26, 2026 | ||
| Salient Object Detection | SOD 300 | 21 | Feb 26, 2026 | ||
| Named Entity Recognition | NEWS | 21 | Feb 27, 2026 | ||
| Amodal instance segmentation | KINS (test) | 21 | May 26, 2026 | ||
| Video Representation | ND Scene (Individual Sequences) | 21 | Feb 26, 2026 | ||
| Image Classification | S-Tiny-ImageNet | 21 | Feb 26, 2026 | ||
| Time Series Imputation | METR-LA Point missing (test) | 21 | Feb 26, 2026 | ||
| Time Series Imputation | PEMS-BAY Point missing (test) | 21 | Feb 26, 2026 | ||
| Time Series Imputation | PEMS-BAY Block missing (test) | 21 | Feb 26, 2026 | ||
| Regression | King (test) | 21 | Feb 26, 2026 | ||
| Long-tailed Image Classification | CIFAR-100-LT λ=100 (test) | 21 | Feb 26, 2026 |