Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 1984 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Low-light Image Enhancement | SICE v2 | 17 | Jul 7, 2026 | ||
| Cell type deconvolution | OOD experiment dataset Out-of-Distribution | 17 | Jul 7, 2026 | ||
| Generated Video Detection | GenVidBench Pair2 (test) | 17 | Jul 7, 2026 | ||
| Synthetic Video Detection | GVF | 17 | Jul 7, 2026 | ||
| Multi-manipulator Path Planning | 3-manipulator simulation environment | 17 | Jul 7, 2026 | ||
| Multi-manipulator planning | Multi-manipulator simulation 5 manipulators | 17 | Jul 7, 2026 | ||
| 3D Object Detection | OmniHD-Scenes (test) | 17 | Jul 7, 2026 | ||
| 3D Object Detection | OmniHD-Scenes night and rainy weather (test) | 17 | Jul 7, 2026 | ||
| Sampling diversity and quality estimation | Gaussian mixture | 17 | Jul 7, 2026 | ||
| Social Navigation Metric Evaluation | SocNav (test) | 17 | Jul 7, 2026 | ||
| Knees Forward Error Detection | BackSquat | 17 | Jul 7, 2026 | ||
| Knees Inward Error Detection | BackSquat | 17 | Jul 7, 2026 | ||
| Image Quality Assessment | Full benchmark (N=144) | 17 | Jul 7, 2026 | ||
| Image Quality Assessment | Coding-only N=80 | 17 | Jul 7, 2026 | ||
| Image Classification | GOOD-CMNIST color | 17 | Jul 7, 2026 | ||
| Semantic Segmentation | Pascal VOC 1/4 labeled split 2012 | 17 | Jul 7, 2026 | ||
| Semantic Segmentation | Pascal VOC 1/8 labeled 2012 | 17 | Jul 7, 2026 | ||
| Semantic Segmentation | Pascal VOC 1/16 labeled 2012 | 17 | Jul 7, 2026 | ||
| In-hospital mortality prediction | MIMIC-IV MedMod protocol (test) | 17 | Jul 7, 2026 | ||
| Human Activity Recognition | HAR (30% train) | 17 | Jul 7, 2026 | ||
| Local prediction | HAR 40% train data (test) | 17 | Jul 7, 2026 | ||
| Visual Question Answering | RoboSpatial-Home | 17 | Jul 7, 2026 | ||
| Discourse Parsing | MSDC | 17 | Jul 3, 2026 | ||
| Discourse Parsing | STAC | 17 | Jul 3, 2026 | ||
| Language Modeling | WikiText-2 (val) | 17 | Jul 3, 2026 |