Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 1332 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Node Classification | Deezer Europe (50% 25% 25% random) | 23 | Feb 26, 2026 | ||
| Action Recognition | NTU-RGB+D 120 1.0 (cross-subject (XSub)) | 23 | Feb 26, 2026 | ||
| Image Enhancement | MIT-Adobe FiveK (Expert C) | 23 | Feb 26, 2026 | ||
| Anomaly Segmentation | Road Anomaly | 23 | Jul 7, 2026 | ||
| Question Answering | Web Questions | 23 | Jun 15, 2026 | ||
| Point Cloud Generation | ShapeNet Chair | 23 | Feb 26, 2026 | ||
| Fine-grained classification | FGVCAircraft Base-to-New | 23 | Feb 26, 2026 | ||
| Hand Manipulation | Adroit door-cloned | 23 | Mar 17, 2026 | ||
| Action Recognition | Ego4D v1 (test) | 23 | Feb 26, 2026 | ||
| Scene Text Detection | CTW1500 (test) | 23 | Feb 26, 2026 | ||
| Speech Enhancement | DNS (test) | 23 | Feb 26, 2026 | ||
| Action Recognition | Kinetics 400/600 (val) | 23 | Feb 26, 2026 | ||
| Action Recognition | UCF-101 1.0 (test) | 23 | Feb 26, 2026 | ||
| 5-way 3-shot Image Classification | MiniImageNet 33.3% symmetric label swap noise 54 (test) | 23 | Feb 26, 2026 | ||
| 5-way 3-shot Image Classification | MiniImageNet 0% noise 54 (test) | 23 | Feb 26, 2026 | ||
| Natural Language Queries | Ego4D NLQ (val) | 23 | Feb 26, 2026 | ||
| Continual Image Classification | ImageNet-100 10 steps | 23 | Feb 26, 2026 | ||
| Video Retrieval | SVD | 23 | Feb 26, 2026 | ||
| Multi-contrast MRI Synthesis (T2, PD -> T1) | IXI (test) | 23 | Feb 26, 2026 | ||
| Audio Classification | UrbanSound8K (official 10 fold split) | 23 | Jun 17, 2026 | ||
| Action Spotting | SoccerNet v2 (test) | 23 | Feb 26, 2026 | ||
| Fine-grained Image Classification | Cars 1.0 (val) | 23 | Feb 26, 2026 | ||
| Question Answering | MRQA Average across 6 domains | 23 | Mar 12, 2026 | ||
| Question Answering | DROP MRQA out-of-domain evaluation | 23 | Mar 12, 2026 | ||
| Question Answering | DuoRC MRQA out-of-domain evaluation | 23 | Mar 12, 2026 |