Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 1326 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Keyphrase Generation | KP20k (test) | 23 | Feb 26, 2026 | ||
| Contamination Detection | WikiMIA | 23 | Feb 26, 2026 | ||
| Instance Segmentation | OV-LVIS | 23 | Mar 19, 2026 | ||
| Event Forecasting | Taxi | 23 | Mar 19, 2026 | ||
| Multi-Classification | LIAR Open | 23 | Feb 26, 2026 | ||
| Binary Classification | LIAR Closed | 23 | Feb 26, 2026 | ||
| Mathematical Reasoning | MMLU STEM | 23 | May 22, 2026 | ||
| Multivariate long-term forecasting | Electricity T=96 (test) | 23 | Feb 26, 2026 | ||
| Commonsense Question Answering | CommonsenseQA (CSQA) (val) | 23 | Feb 26, 2026 | ||
| Grasp Detection | GraspNet-1Billion Kinect camera (seen) | 23 | Feb 26, 2026 | ||
| Automatic Speech Recognition | CORAAL (test) | 23 | Jun 11, 2026 | ||
| Image Classification | DigitsDG | 23 | May 29, 2026 | ||
| Human-human interaction motion generation | InterHuman | 23 | Feb 26, 2026 | ||
| Open Set Domain Generalization | OfficeHome H=1 | 23 | Feb 26, 2026 | ||
| Sound Target Segmentation | AVSBench-object MS3 1.0 (test) | 23 | Feb 26, 2026 | ||
| Video Classification | Something-Something V1 | 23 | Apr 23, 2026 | ||
| 3D Object Detection | ARKitScenes | 23 | Mar 24, 2026 | ||
| Person Re-identification | AG-ReID (A -> C) v2 (test) | 23 | Feb 26, 2026 | ||
| Continuous Sign Language Recognition | PHOENIX14 2015 (dev) | 23 | Apr 21, 2026 | ||
| Blind Video Quality Assessment | LIVE-YT-HFR | 23 | May 19, 2026 | ||
| Visual Object Tracking | GOT-10k Restricted Protocol (test) | 23 | Feb 26, 2026 | ||
| Image Classification | SVHN (train) | 23 | Apr 27, 2026 | ||
| Semantic Segmentation | COCO-Object with background class (val) | 23 | Feb 26, 2026 | ||
| Image Classification | ImageWoof full-sized (test) | 23 | Feb 26, 2026 | ||
| LiDAR Scene Flow Estimation | Argoverse v2 (val) | 23 | Feb 26, 2026 |