Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 1647 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Video Instance Parsing | VIP (val) | 20 | Feb 26, 2026 | ||
| Indoor Semantic Segmentation | ScanNet++ (val) | 20 | Mar 4, 2026 | ||
| Fair Generation | WinoBias Race-Pro (extended) | 20 | Feb 26, 2026 | ||
| Fair Generation | WinoBias Race (standard) | 20 | Feb 26, 2026 | ||
| Fair Generation | WinoBias Gender-Pro extended | 20 | Feb 26, 2026 | ||
| Fair Generation | WinoBias Gender (standard) | 20 | Feb 26, 2026 | ||
| Histopathological Image Segmentation | BCSS-WSSS | 20 | Feb 26, 2026 | ||
| Video Classification | COIN (test) | 20 | Feb 26, 2026 | ||
| Video Person Re-identification | AG-VPReID Aerial to Ground | 20 | Feb 26, 2026 | ||
| Dialogue Generation | Douban (test) | 20 | Feb 26, 2026 | ||
| Camera Pose Localization | Cambridge Landmarks | 20 | Mar 4, 2026 | ||
| Mathematical Reasoning | GSM8k | 20 | Feb 26, 2026 | ||
| Classification | TinyImageNet | 20 | May 18, 2026 | ||
| Domain Verification | Office-31 | 20 | Feb 26, 2026 | ||
| Image Segmentation | BIG (test) | 20 | Feb 26, 2026 | ||
| Conversational versatility | Arena-Hard | 20 | Feb 26, 2026 | ||
| Spatio-Temporal Video Grounding | VidSTG Declarative Sentences | 20 | Feb 26, 2026 | ||
| Visual Question Answering | SLAKE (test) | 20 | Mar 30, 2026 | ||
| Classification | HELOC | 20 | Feb 26, 2026 | ||
| Building Extraction | WHU (test) | 20 | Feb 26, 2026 | ||
| Language Modeling | The Pile non-copyrighted (test) | 20 | Feb 26, 2026 | ||
| Class-Incremental Learning | Tiny-ImageNet standard (test) | 20 | Feb 26, 2026 | ||
| Image Classification | VLCS DomainBed suite (test) | 20 | Feb 26, 2026 | ||
| Image Classification | PACS DomainBed suite (test) | 20 | Feb 26, 2026 | ||
| Time Series Forecasting | FaaS | 20 | Feb 26, 2026 |