Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 601 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Relation Extraction | DocRED official (dev) | 38 | Feb 26, 2026 | ||
| Capacitated Vehicle Routing Problem | CVRP N=20 10,000 instances (test) | 38 | May 21, 2026 | ||
| Image Classification | MNIST (test) | 38 | Feb 26, 2026 | ||
| Saliency Detection | DUT-OMRON 29 (test) | 38 | Feb 26, 2026 | ||
| Speaker Verification | CNCeleb | 38 | Jul 7, 2026 | ||
| Offline Reinforcement Learning | D4RL Walker2D Expert | 38 | May 11, 2026 | ||
| General Robust Image Task (GRIT) multi-task evaluation | GRIT ablation set (same) | 38 | Feb 26, 2026 | ||
| Image Classification | PC-MNIST (test) | 38 | Feb 26, 2026 | ||
| Image Classification | PC-MNIST (val) | 38 | Feb 26, 2026 | ||
| Image Generation | ImageNet 1K (train) | 38 | Mar 9, 2026 | ||
| Image Harmonization | iHarmony4 HCOCO | 38 | Feb 26, 2026 | ||
| Depth Prediction | KITTI original ground truth (test) | 38 | Feb 26, 2026 | ||
| Object Detection | PASCAL VOC 2007 (test) | 38 | Feb 27, 2026 | ||
| No-Reference Image Quality Assessment | SPAQ (test) | 38 | Mar 25, 2026 | ||
| Motion Forecasting | Argoverse 1.0 (val) | 38 | Mar 31, 2026 | ||
| Point Cloud Quality Assessment | SJTU-PCQA | 38 | Feb 26, 2026 | ||
| Video prediction | BAIR Robot Pushing | 38 | Feb 26, 2026 | ||
| Semantic Segmentation | Cityscapes (val) | 38 | Feb 26, 2026 | ||
| Code Summarization | CodeSearchNet-Java (CSN) CodeXGLUE (test) | 38 | Feb 26, 2026 | ||
| Space-Time Video Super-Resolution | Adobe-Average (test) | 38 | Apr 21, 2026 | ||
| LF Mislabeling Identification | IMDb | 38 | Feb 26, 2026 | ||
| Few-shot Classification | tiered-ImageNet | 38 | Feb 26, 2026 | ||
| Few-shot Action Recognition | SS Full meta v2 (test) | 38 | Feb 26, 2026 | ||
| Audio-visual Zero-Shot Classification | VGGSound GZSL (test) | 38 | Feb 27, 2026 | ||
| Test-time adaptation | 7 scenarios (ImageNet-C, ImageNet-V2, ImageNet-VID, LaSOT subset from TAO) (test) | 38 | Feb 26, 2026 |