Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 1476 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Visual Question Answering | ScienceQA SQA-I | 21 | Jun 23, 2026 | ||
| Road segment ranking | SY-Net514 (test) | 21 | Feb 26, 2026 | ||
| Road Segment Ranking | RD-Net3478 (test) | 21 | Feb 26, 2026 | ||
| Road segment ranking | SY-Net110 | 21 | Feb 26, 2026 | ||
| Commonsense and Mathematical Reasoning | Reasoning and Math Suite (OpenBookQA, ARC-e, WinoGrande, HellaSwag, ARC-c, PIQA, MathQA, GSM8K) | 21 | Feb 26, 2026 | ||
| Clustering | BC | 21 | Jun 23, 2026 | ||
| Multi-modal brain MRI synthesis | BraTS 2021 (test) | 21 | Feb 26, 2026 | ||
| Medical Image Classification | VesselMNIST3D | 21 | May 27, 2026 | ||
| Medical Image Classification | AdrenalMNIST3D | 21 | May 27, 2026 | ||
| Mathematical Reasoning | AIME 25 | 21 | Mar 31, 2026 | ||
| Video Person Re-Identification | Market-1501 v1 (test) | 21 | Feb 26, 2026 | ||
| Few-shot Action Recognition | UCF101 5-way 1-shot | 21 | Feb 26, 2026 | ||
| Classification | heart (test) | 21 | Apr 10, 2026 | ||
| Node Classification | Twitch | 21 | Feb 26, 2026 | ||
| Concept Learning | AG News | 21 | Feb 26, 2026 | ||
| Reasoning | 17 Reasoning Benchmarks Aggregate (test) | 21 | Feb 26, 2026 | ||
| Time Series Classification | MP8 | 21 | Jul 1, 2026 | ||
| Time Series Classification | HHAR-OOD | 21 | Jul 1, 2026 | ||
| Time Series Classification | Epilepsy-EEG | 21 | Jul 1, 2026 | ||
| Out-of-Distribution Detection | CIFAR-10 LT (ρ = 0.1) | 21 | May 26, 2026 | ||
| Person Re-identification | CUHK03-NP (labeled) | 21 | Feb 26, 2026 | ||
| Virtual Try-on | VITON-HD 1024x768 (test) | 21 | Jun 29, 2026 | ||
| Point Cloud Registration | ModelNet40 Unseen Categories with Gaussian Noise RPMNet manner (OS) | 21 | Feb 26, 2026 | ||
| Depth Completion | NYU v2 (test) | 21 | May 29, 2026 | ||
| Speech deepfake detection | ASVspoof logical access (LA) 2019 (eval) | 21 | Feb 26, 2026 |