Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 1240 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Novel View Synthesis | Panoptic Studio CMU (test) | 24 | Feb 26, 2026 | ||
| General Video Understanding | LongVideoBench | 24 | Jun 23, 2026 | ||
| SIRST Detection | SIRST3 | 24 | Feb 26, 2026 | ||
| Object Hallucination | COCO 512-token budget (test) | 24 | Feb 26, 2026 | ||
| Object Hallucination | COCO 64-token budget (test) | 24 | Feb 26, 2026 | ||
| AI-generated image detection | StyleGAN2 | 24 | Mar 4, 2026 | ||
| AI-generated image detection | StarGAN | 24 | Mar 4, 2026 | ||
| Monocular 3D Object Detection | Waymo Open Dataset 79 (val) | 24 | Feb 26, 2026 | ||
| Infant respiration estimation | AIR-400 1.0 (six-fold subject-wise cross-val) | 24 | Feb 26, 2026 | ||
| Closed-loop Planning | NAVSIM v2 (test) | 24 | May 20, 2026 | ||
| Camera Pose Estimation | Sintel 14-sequence | 24 | Mar 4, 2026 | ||
| Object Visual Question Answering (Open Ended) | RIO-Bench Obj-VQA (OE) Attack | 24 | Feb 26, 2026 | ||
| Object Visual Question Answering (Open Ended) | RIO-Bench Obj-VQA (OE) Clean | 24 | Feb 26, 2026 | ||
| Hierarchical Unlearning | MedForget 1.0 (Retain) | 24 | Feb 26, 2026 | ||
| Image Classification | 16 single-label datasets Aggregate | 24 | Feb 27, 2026 | ||
| Novel View Synthesis | Custom | 24 | Feb 26, 2026 | ||
| Image Classification | Target Classes Forget Set (test) | 24 | Feb 27, 2026 | ||
| Window-level anti-pattern detection | TorchTraceAP (Unseen) | 24 | Feb 26, 2026 | ||
| Image Classification | PASCAL VOC | 24 | Jun 11, 2026 | ||
| Image-to-Multi-RGBA | Crello (test) | 24 | Feb 26, 2026 | ||
| Multimodal Understanding | MMMU-Pro std. | 24 | Apr 29, 2026 | ||
| 3D Computer Vision Benchmarking | CVBench3D | 24 | May 25, 2026 | ||
| Vehicle Dispatching for Map Reconstruction | Mobile Crowdsensing Air Quality Simulation | 24 | Feb 26, 2026 | ||
| Super-Resolution | Flickr360 3 (val) | 24 | Feb 26, 2026 | ||
| Deepfake Detection | Aggregate All Unseen (test) | 24 | Feb 26, 2026 |