Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 1869 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Defense against adaptive adversarial attacks | CelebA | 18 | Mar 24, 2026 | ||
| Defense against adaptive adversarial attacks | Bank | 18 | Mar 24, 2026 | ||
| Mathematical Reasoning | GSM8K | 18 | Mar 24, 2026 | ||
| 1D Regression | 1D Regression 20 points (test) | 18 | Mar 24, 2026 | ||
| MaxCut | GSET (|V|>=3K) | 18 | May 26, 2026 | ||
| MaxCut | GSET |V|=2K | 18 | May 26, 2026 | ||
| No-reference Image Quality Assessment | AGIQA | 18 | Mar 24, 2026 | ||
| Edge Detection | BSDS (test) | 18 | Mar 24, 2026 | ||
| Agentic Coding | Terminal Bench 2.0 | 18 | May 28, 2026 | ||
| Video Understanding | TOMATO | 18 | Mar 31, 2026 | ||
| Text-to-Image Generation | LAION 5B 1K | 18 | Mar 24, 2026 | ||
| Text-to-Image Generation | MS-COCO 1K | 18 | Mar 24, 2026 | ||
| End-to-end Retrieval | MSMARCO | 18 | Mar 24, 2026 | ||
| Scientific Paper Retrieval | SCIFULLBENCH ACL-EMNLP (Citations) 1.0 | 18 | Mar 24, 2026 | ||
| Scientific Paper Retrieval | SCIFULLBENCH ACL-EMNLP (References) 1.0 | 18 | Mar 24, 2026 | ||
| Scientific Paper Retrieval | SCIFULLBENCH ICLR-NeurIPS (Citations) 1.0 | 18 | Mar 24, 2026 | ||
| Scientific Paper Retrieval | SCIFULLBENCH ICLR-NeurIPS (References) 1.0 | 18 | Mar 24, 2026 | ||
| Trajectory Prediction | Simulated Dataset | 18 | Mar 24, 2026 | ||
| Trajectory Prediction | NGSIM | 18 | Mar 24, 2026 | ||
| Diverse Image Segmentation | LIDC | 18 | Mar 23, 2026 | ||
| Diverse Image Segmentation | Cityscapes | 18 | Mar 23, 2026 | ||
| Value Function Estimation | Game B Matrix Game | 18 | Mar 23, 2026 | ||
| Sentiment Classification | CM, EN, and HI (test) | 18 | Mar 23, 2026 | ||
| Sentiment Classification | Hindi (HI) (test) | 18 | Mar 23, 2026 | ||
| Sentiment Classification | English (EN) (test) | 18 | Mar 23, 2026 |