Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 828 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Classification | Ten-dimensional normal data (d=10) (test) | 30 | Jun 23, 2026 | ||
| System Prompt Corruption (SPC) | COMA Evaluation Suite | 30 | Jun 23, 2026 | ||
| Question Answering (Q&A) | COMA Evaluation Suite | 30 | Jun 23, 2026 | ||
| Adversarial Task Switching (ATS) | COMA Evaluation Suite | 30 | Jun 23, 2026 | ||
| Image Forgery Localization | Columbia Unseen Domain | 30 | Jun 23, 2026 | ||
| Anomaly Detection | MVTec AD (test) | 30 | Jun 23, 2026 | ||
| Adversarial Attack Success | AdvBench (sample of 500) | 30 | Jun 19, 2026 | ||
| 3D Semantic Segmentation | Waymo | 30 | Jun 19, 2026 | ||
| 3D Shape Generation | 3D Shape Generation Dataset (test) | 30 | Jun 19, 2026 | ||
| LiDAR-Camera Calibration | KITTI-360 | 30 | Jun 19, 2026 | ||
| Pairwise Comparison Accuracy Refinement | Coding | 30 | Jun 19, 2026 | ||
| Coding | Coding Real-data 20% verified | 30 | Jun 19, 2026 | ||
| Coding | LiveCodeBench Jan 25 – May 25 | 30 | Jun 18, 2026 | ||
| Strategic Action Utility Evaluation | Real-world liquidity provisioning game | 30 | Jun 18, 2026 | ||
| World modelling | ScienceWorld | 30 | Jun 17, 2026 | ||
| World Modeling | ScienceWorld | 30 | Jun 17, 2026 | ||
| Image Deblurring | GoPro CosPGD attack 10 iterations | 30 | Jun 17, 2026 | ||
| Cough-based Tuberculosis Screening | CODA (test) | 30 | Jun 17, 2026 | ||
| Multimodal Evaluation | MME-R | 30 | Jun 17, 2026 | ||
| Open-vocabulary semantic segmentation | COCO Object | 30 | Jun 16, 2026 | ||
| Inpainting | AFHQ cats | 30 | Jun 16, 2026 | ||
| Text-to-SQL Error Repairing | ScienceBenchmark | 30 | Jun 16, 2026 | ||
| Text-to-SQL Error Repairing | Spider | 30 | Jun 16, 2026 | ||
| Text-to-SQL Error Repairing | BIRD | 30 | Jun 16, 2026 | ||
| Neuroretinal Rim Segmentation | ORIGA (OOD) | 30 | Jun 16, 2026 |