Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 1254 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Reasoning | Word Sorting | 24 | Feb 26, 2026 | ||
| Jailbreak Defense | Jailbreak Attack Benchmarks (GPTFuzz, TAP, GCG, AutoDAN, Template) | 24 | Feb 26, 2026 | ||
| Safety Evaluation | Harmful Benchmarks (CATQA, HEX-PHI, Salad-Base) | 24 | Feb 26, 2026 | ||
| Surgical Instrument Segmentation | Robust-MIS 2019 (test) | 24 | Feb 26, 2026 | ||
| Jailbreak Attack | Jailbreak prompts Manufacturing and distributing illegal drugs | 24 | Feb 26, 2026 | ||
| Factual Question Answering | LSQA OOD | 24 | Feb 26, 2026 | ||
| Factual Question Answering | ID Datasets Average | 24 | Feb 26, 2026 | ||
| Factual Question Answering | NQ-Open ID | 24 | Feb 26, 2026 | ||
| Factual Question Answering | SciQ (ID) | 24 | Feb 26, 2026 | ||
| Factual Question Answering | TVQA ID | 24 | Feb 26, 2026 | ||
| Object Detection | PASCAL VOC Set 3 | 24 | Feb 27, 2026 | ||
| Object Detection | PASCAL VOC Set 2 | 24 | Feb 27, 2026 | ||
| Object Detection | PASCAL VOC Set 1 | 24 | Feb 27, 2026 | ||
| Information Retrieval | SCIDocs | 24 | Feb 26, 2026 | ||
| Re-identification | TAB (Text Anonymization Benchmark) | 24 | Feb 26, 2026 | ||
| Machine-Generated Text Detection | MAGE | 24 | May 11, 2026 | ||
| Polyp Segmentation | Overall Combined 5 Datasets (test) | 24 | Feb 26, 2026 | ||
| Continuous Space-Time Video Super-Resolution | Standard Clip 180x320 resolution | 24 | Feb 26, 2026 | ||
| Document and chart understanding | CharXiv DQ | 24 | Jul 2, 2026 | ||
| Stereo Matching | Middlebury 2021 | 24 | Mar 4, 2026 | ||
| Tile-level classification | PatchCamelyon | 24 | Feb 26, 2026 | ||
| Trajectory Prediction | SDD (test) | 24 | Apr 28, 2026 | ||
| Zero-shot Audio Classification Explanation | ESC50 contamination | 24 | Feb 26, 2026 | ||
| Image Classification | RSNA | 24 | Feb 26, 2026 | ||
| Long-term forecasting | ECL T=192 | 24 | May 15, 2026 |