Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 347 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Long Form Question Answering | ELI5 (test) | 54 | Feb 27, 2026 | ||
| Visual Commonsense Reasoning | VCR (Visual Commonsense Reasoning) (test) | 54 | Feb 26, 2026 | ||
| Video Inpainting | DAVIS (test) | 54 | Feb 26, 2026 | ||
| Dense Video Captioning | ActivityNet Captions (val) | 54 | Feb 27, 2026 | ||
| Document classification | Ohsumed (test) | 54 | Feb 26, 2026 | ||
| Landmark Localization | AFLW (test) | 54 | Feb 26, 2026 | ||
| Summarization | CNN/Daily Mail original, non-anonymized (test) | 54 | Feb 26, 2026 | ||
| Multivariate Time Series Forecasting | Electricity (test) | 54 | May 11, 2026 | ||
| Face Verification | CFP (Frontal-Frontal) | 54 | Feb 26, 2026 | ||
| Weakly Supervised Object Localization | PASCAL VOC 2007 (trainval) | 54 | Feb 26, 2026 | ||
| Semantic Segmentation | PASCAL-Context 60 classes (test) | 54 | Feb 26, 2026 | ||
| Object Detection | MS COCO 2015 (test-dev) | 54 | Feb 26, 2026 | ||
| Unconditional Molecule Generation | QM9 (test) | 54 | May 21, 2026 | ||
| Question Answering | bAbI (test) | 54 | Feb 26, 2026 | ||
| Regression | CA Housing | 54 | May 19, 2026 | ||
| Image Classification | Tiny ImageNet (Tiny-IN) (val) | 54 | Feb 26, 2026 | ||
| Sleep Stage Classification | SHHS (test) | 54 | Feb 26, 2026 | ||
| Text-to-Image Generation | MSCOCO 30K | 54 | Mar 10, 2026 | ||
| Offline Reinforcement Learning | halfcheetah medium-replay | 54 | May 29, 2026 | ||
| Class-incremental learning | ImageNet-R 10-task | 54 | May 19, 2026 | ||
| Arithmetic Reasoning | SVAMP | 54 | Feb 26, 2026 | ||
| WSI Classification | BRACS (test) | 54 | Jun 23, 2026 | ||
| Anomaly Classification | OCT 17 | 54 | Feb 26, 2026 | ||
| Medical Visual Question Answering | SLAKE closed-end | 54 | Feb 26, 2026 | ||
| Survival Prediction | TCGA-BRCA (5-fold cross-validation) | 54 | Feb 26, 2026 |