Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 1978 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Zero-shot classification | AwA 10-way 0-shot conventional setting | 18 | Feb 26, 2026 | ||
| Ground-to-Aerial Image Synthesis | Dayton 64x64 | 18 | Feb 26, 2026 | ||
| Aerial-to-Ground Image Synthesis | Dayton 64x64 | 18 | Feb 26, 2026 | ||
| 11-class movement classification | EEG-EMG Inter-subject | 18 | Feb 26, 2026 | ||
| Human Parsing | CIHP | 18 | Feb 26, 2026 | ||
| Multi-label Classification | MS-COCO 2014 (val) | 18 | Feb 26, 2026 | ||
| Facial Beauty Prediction | SCUT-FBP | 18 | Mar 19, 2026 | ||
| Local Descriptor Matching | 3DMatch 1.0 (test) | 18 | Feb 26, 2026 | ||
| Question Answering | SQuAD hidden 1.1 (test) | 18 | Feb 26, 2026 | ||
| Action Recognition | JIGSAWS Suturing (LOSO) | 18 | Feb 26, 2026 | ||
| Gender Recognition | CelebA (test) | 18 | Feb 26, 2026 | ||
| Text-to-Image Generation | COCO (test) | 18 | Apr 20, 2026 | ||
| pixel-by-pixel classification | MNIST unpermuted pixel-by-pixel (test) | 18 | Feb 26, 2026 | ||
| Gender Classification | Adience (cross-validation) | 18 | Feb 26, 2026 | ||
| Salient Object Detection | SOD | 18 | Feb 26, 2026 | ||
| Salient Object Detection | DUTS (TE) | 18 | Feb 26, 2026 | ||
| Natural Language Inference | MultiNLI matched (in-domain) RepEval 2017 (test) | 18 | Feb 26, 2026 | ||
| 3D Human Pose Estimation | HumanEva-I (Walking) | 18 | Feb 26, 2026 | ||
| Shot Boundary Detection | TRECVID 2007 (test) | 18 | Feb 26, 2026 | ||
| 20-way Few-shot Classification | Omniglot (test) | 18 | Feb 26, 2026 | ||
| Generative Question Answering | MsMARCO (test) | 18 | Feb 26, 2026 | ||
| Individual Treatment Effect Estimation | Jobs (within-sample) | 18 | Feb 26, 2026 | ||
| Semantic Segmentation | Semantic3D reduced v1 (test) | 18 | Feb 26, 2026 | ||
| Short Text Clustering | Biomedical | 18 | Feb 26, 2026 | ||
| Individual Action Recognition | Volleyball dataset | 18 | Feb 27, 2026 |