Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 792 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Entity Translation | WMT newstest (test) | 32 | Feb 26, 2026 | ||
| Relation Extraction | BioRED (test) | 32 | Feb 26, 2026 | ||
| Adversarial Attack | GLUE | 32 | Feb 26, 2026 | ||
| Natural Language Understanding | GLUE | 32 | Feb 26, 2026 | ||
| Heat Equation Simulation | 2D battery pack | 32 | Feb 27, 2026 | ||
| Text Classification | FCDL18 | 32 | Feb 26, 2026 | ||
| Top-level Implicit Discourse Relation Recognition | PDTB Lin split 2.0 | 32 | Feb 26, 2026 | ||
| Reading Comprehension | Quoref | 32 | Apr 10, 2026 | ||
| Semantic Segmentation | Stanford2D3D | 32 | Feb 26, 2026 | ||
| Video Anomaly Detection | Ped2 classic (test) | 32 | May 26, 2026 | ||
| Out-of-Distribution Detection | Aggregated OOD benchmark 308 experiments (test) | 32 | Feb 26, 2026 | ||
| Video Prediction | UCF Sports t+1 (test) | 32 | Feb 26, 2026 | ||
| Image Denoising | Kodak24 Mixture noise | 32 | Feb 26, 2026 | ||
| Image Denoising | Kodak24 Salt & pepper noise | 32 | Feb 26, 2026 | ||
| Image Denoising | Kodak24 Spatially-correlated noise | 32 | Feb 26, 2026 | ||
| Image Denoising | Kodak24 Poisson noise | 32 | Feb 26, 2026 | ||
| Image Denoising | Kodak24 Speckle noise | 32 | Feb 26, 2026 | ||
| Image Denoising (Mixture Noise) | Urban100 (test) | 32 | Feb 26, 2026 | ||
| Image Denoising (Salt & Pepper Noise) | Urban100 (test) | 32 | Feb 26, 2026 | ||
| Image Denoising (Spatially-correlated Noise) | Urban100 (test) | 32 | Feb 26, 2026 | ||
| Image Denoising (Poisson Noise) | Urban100 (test) | 32 | Feb 26, 2026 | ||
| Image Denoising (Speckle Noise) | Urban100 (test) | 32 | Feb 26, 2026 | ||
| Long-Tail Image Classification | CIFAR100-LT imbalance ratio 100 (test) | 32 | Feb 27, 2026 | ||
| 3D Human Pose Estimation | CMU Panoptic | 32 | May 12, 2026 | ||
| Traffic Forecasting | PEMS-BAY 15min horizon 3 | 32 | May 8, 2026 |