Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 389 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Object Detection | ARCADE real (test) | 49 | May 12, 2026 | ||
| Novel View Synthesis | V2X-Seq (All sequences) | 49 | May 11, 2026 | ||
| Scene Reconstruction | V2X-Seq (All sequences) | 49 | May 11, 2026 | ||
| Claim Verification | Strategy Claim | 49 | May 8, 2026 | ||
| Claim Verification | Truthful Claim | 49 | May 8, 2026 | ||
| Watermark Robustness Analysis | Gemma-2 2B | 49 | May 8, 2026 | ||
| Watermarking Attack Robustness | Gemma 9B v2 (test) | 49 | May 8, 2026 | ||
| Knowledge Retention | Knowledge Retention Evaluation Set | 49 | May 7, 2026 | ||
| Backdoor Detection | EMNIST | 49 | May 7, 2026 | ||
| Mathematical Reasoning | MathVerse V | 49 | Jun 23, 2026 | ||
| Harmful Content Detection | HoliSafe-Bench | 49 | May 4, 2026 | ||
| Safety Classification | HoliSafe-Bench | 49 | May 4, 2026 | ||
| Safety Classification | UnsafeBench | 49 | May 4, 2026 | ||
| Skin Lesion Segmentation | ISIC 2016 | 49 | Jul 2, 2026 | ||
| Continual Supervised Learning | Continual ImageNet | 49 | May 18, 2026 | ||
| Continual Supervised Learning | CIFAR 5+1 | 49 | May 18, 2026 | ||
| Continual Supervised Learning | CIFAR Random Label | 49 | May 18, 2026 | ||
| Image Classification | ImageNet-1K (val) | 49 | Apr 29, 2026 | ||
| Instruction Following | IFEval | 49 | Jun 26, 2026 | ||
| Backdoor Detection | CIFAR-100 | 49 | May 7, 2026 | ||
| Biosecurity Misuse Evaluation | BSD Biosecurity | 49 | Apr 23, 2026 | ||
| Math Reasoning | GSM8K | 49 | Apr 23, 2026 | ||
| Visual Localization | i7Scenes | 49 | Apr 20, 2026 | ||
| Drone-to-Satellite Cross-view Geo-localization | SUES-200 250m | 49 | May 12, 2026 | ||
| Pose Estimation | ETH3D | 49 | Jul 7, 2026 |