Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 1663 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Semantic Segmentation | DALES (test) | 20 | Mar 25, 2026 | ||
| Occupation classification | Occupation classification balanced (test) | 20 | Feb 26, 2026 | ||
| Occupation classification | Occupation classification dataset | 20 | Feb 26, 2026 | ||
| Object Detection | VOC to Comic (test) | 20 | Feb 26, 2026 | ||
| Model Inversion | CelebA | 20 | Jun 23, 2026 | ||
| Medical Image Segmentation | KITS | 20 | Apr 9, 2026 | ||
| Machine Translation | newstest 18 (head150) | 20 | Feb 26, 2026 | ||
| Retrieval | Entity Questions (test) | 20 | Feb 26, 2026 | ||
| Hierarchical Text Classification | RCV1 Depth 4 V2 | 20 | Feb 27, 2026 | ||
| Pixel-level Forgery Localization | Columbia | 20 | Apr 15, 2026 | ||
| Image Forgery Localization | CASIA v1 | 20 | Jun 23, 2026 | ||
| Image Forgery Detection | CocoGlide | 20 | Jun 11, 2026 | ||
| Full-Reference Image Quality Assessment | LIVE-IQA | 20 | Feb 26, 2026 | ||
| Controllable Text Generation | Yelp (test) | 20 | Feb 26, 2026 | ||
| Question Answering | NewsQA trained on SQuAD OOD (test) | 20 | Feb 26, 2026 | ||
| Few-Shot Open-Set Recognition | tiered-ImageNet 5-shot | 20 | Feb 26, 2026 | ||
| Few-Shot Open-Set Recognition | tiered-ImageNet 1-shot | 20 | Feb 26, 2026 | ||
| Few-Shot Open-Set Recognition | mini-ImageNet 5-shot | 20 | Feb 26, 2026 | ||
| Few-Shot Open-Set Recognition | mini-ImageNet 1-shot | 20 | Feb 26, 2026 | ||
| Semantic Segmentation | DeepGlobe (test) | 20 | Feb 27, 2026 | ||
| Argument Identification | WikiEvents (test) | 20 | Feb 26, 2026 | ||
| Multi-task Evaluation | Aggregate All tasks (summary) | 20 | Feb 26, 2026 | ||
| 3D Object Detection | S3DIS (val) | 20 | Feb 26, 2026 | ||
| Novel View Synthesis | LLFF Forward-facing (test) | 20 | Feb 26, 2026 | ||
| Factual Consistency Evaluation | TRUE 1.0 (test) | 20 | Feb 26, 2026 |