Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 1650 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Image Fusion | Harvard Medicine Dataset (test) | 20 | Mar 25, 2026 | ||
| Class Conditional Image Generation | CIFAR-10 class conditional 32x32 | 20 | Feb 26, 2026 | ||
| Time Series Forecasting | ETTh1 -> ETTm1 | 20 | Mar 17, 2026 | ||
| Molecule Generation | QM9 (test) | 20 | Mar 6, 2026 | ||
| Segmentation | ShapeNet-C (test) | 20 | Feb 26, 2026 | ||
| Object Detection | PKU-DDD Car 17 | 20 | Feb 26, 2026 | ||
| Offline Policy Evaluation | ToyGraph (test) | 20 | Feb 26, 2026 | ||
| Image Deraining & Dehazing | Outdoor-Rain (test) | 20 | Apr 9, 2026 | ||
| Reward Modeling | RewardBench unified-feedback (test) | 20 | Feb 26, 2026 | ||
| Symbolic Reasoning (Equations) | NoRa Inaccurate Rationales | 20 | Feb 26, 2026 | ||
| Mathematical Reasoning (Base-9) | NoRa Inaccurate Rationales | 20 | Feb 26, 2026 | ||
| Symbolic Reasoning (Equations) | NoRa Irrelevant Rationales | 20 | Feb 26, 2026 | ||
| Mathematical Reasoning | Math Base-9 | 20 | Feb 26, 2026 | ||
| Anomaly Detection | MVTec3D | 20 | Feb 26, 2026 | ||
| Human Activity Recognition | HAPT | 20 | May 20, 2026 | ||
| Activity Recognition | DSADS | 20 | May 22, 2026 | ||
| Activity Recognition | DG | 20 | May 20, 2026 | ||
| Image Restoration | RealSR (test) | 20 | Apr 28, 2026 | ||
| Tabular Classification | TabZilla avg across 98 datasets | 20 | Feb 26, 2026 | ||
| Scene Text-Centric Visual Question Answering | STVQA | 20 | May 21, 2026 | ||
| Link Prediction | Pubmed | 20 | Mar 10, 2026 | ||
| Citation Recommendation | Scientific Paper Domains Overall (test) | 20 | Feb 26, 2026 | ||
| Citation Recommendation | Scientific Paper Domains Social science (test) | 20 | Feb 26, 2026 | ||
| Citation Recommendation | Scientific Paper Domains Natural science (test) | 20 | Feb 26, 2026 | ||
| Folding | VR-Folding 1.0 (test) | 20 | Feb 26, 2026 |