Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 1667 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Image Generation | SVHN (test) | 20 | Mar 31, 2026 | ||
| Few-shot Segmentation | BCV Spleen in-domain (novel classes) | 20 | Feb 26, 2026 | ||
| Multi-task Classification | ImageCLEF | 20 | Feb 26, 2026 | ||
| Multi-task Classification | Office-Caltech | 20 | Feb 26, 2026 | ||
| Multi-task Classification | Office-Home | 20 | Feb 26, 2026 | ||
| Vertex Classification | Mushroom (test) | 20 | Feb 26, 2026 | ||
| Vertex Classification | DBLP-CA (test) | 20 | Feb 26, 2026 | ||
| Vertex Classification | Cora-CA (test) | 20 | Feb 26, 2026 | ||
| Unsupervised Semantic Segmentation | COCO | 20 | Feb 26, 2026 | ||
| Image Recognition | iDigits NC v1 | 20 | Feb 26, 2026 | ||
| Image Recognition | iDomainNet v1 (NC) | 20 | Feb 26, 2026 | ||
| Scene-level 3D Reconstruction | ScanNet (test) | 20 | Feb 26, 2026 | ||
| Multilingual Question Answering | TyDiQA GoldP (val) | 20 | Apr 7, 2026 | ||
| Temporal Knowledge Graph Reasoning | YAGO | 20 | Feb 26, 2026 | ||
| Semantic Segmentation | SemanticKITTI multiple scans (test) | 20 | Feb 26, 2026 | ||
| Natural Language Inference | QNLI 64 instances (test) | 20 | Feb 26, 2026 | ||
| Sentence-pair classification | QNLI | 20 | Feb 26, 2026 | ||
| Trajectory Prediction | TrajAir (7Days-3) | 20 | Mar 18, 2026 | ||
| Visual Sound Source Localization | VGG-SS extended (test) | 20 | Mar 25, 2026 | ||
| Visual Sound Source Localization | Flickr-SoundNet extended (test) | 20 | Mar 25, 2026 | ||
| Fine-grained Image Classification | Cars | 20 | Feb 26, 2026 | ||
| Out-of-Distribution Detection | ImageNet-1K (val) | 20 | Apr 28, 2026 | ||
| Paraphrase Identification | QQP Out-of-distribution from PAWS | 20 | Feb 26, 2026 | ||
| Paraphrase Identification | PAWS -> QQP (test) | 20 | Feb 26, 2026 | ||
| Paraphrase Identification | PIT -> PAWS (test) | 20 | Feb 26, 2026 |