Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 1388 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Micro-gesture classification | iMiGUE | 22 | Mar 18, 2026 | ||
| Relative Pose Estimation | SUN3D Unknown Scene | 22 | Feb 26, 2026 | ||
| Relative Pose Estimation | YFCC100M Unknown Scene | 22 | Feb 26, 2026 | ||
| Intrinsic Decomposition | HyperSim | 22 | Jun 30, 2026 | ||
| Tracking and Mapping | 7Scenes | 22 | May 20, 2026 | ||
| Animation Video Colorization | SAKUGA-42M (test) | 22 | Feb 26, 2026 | ||
| Tactile Recognition | Tactile Cross-Domain OF Real to X Unseen target domains | 22 | Mar 10, 2026 | ||
| Deepfake Detection | FaceForensics++ (FF) (test) | 22 | Feb 26, 2026 | ||
| Temporal Action Segmentation | 50 Salads 65 | 22 | Feb 26, 2026 | ||
| Generative Hallucination | AMBER generative subset | 22 | Feb 26, 2026 | ||
| Object Hallucination | MME | 22 | Feb 26, 2026 | ||
| Object Hallucination | POPE average across COCO, A-OKVQA, GQA | 22 | Feb 26, 2026 | ||
| Multi-concept Customization | ImageNet (INet) (test) | 22 | Feb 26, 2026 | ||
| Multi-concept Customization | CelebA (test) | 22 | Feb 26, 2026 | ||
| Multi-concept Customization | CIFC (test) | 22 | Feb 26, 2026 | ||
| Deepfake Detection | IDForge (test) | 22 | Feb 26, 2026 | ||
| Multi-image understanding | MIBench | 22 | Mar 10, 2026 | ||
| Medical Image-to-Image Translation (T2→FLAIR) | BraTS 2023 (test) | 22 | Jun 1, 2026 | ||
| Video Summarization | Video-MME 900 videos | 22 | Feb 26, 2026 | ||
| Color Constancy | Gehler-Shi | 22 | May 20, 2026 | ||
| Pixel-level Anomaly Detection | Real-IAD | 22 | Feb 26, 2026 | ||
| Image-level Anomaly Detection | Real-IAD | 22 | Feb 26, 2026 | ||
| AI-generated image detection | DDPM generator (test) | 22 | Feb 26, 2026 | ||
| AI-generated image detection | BigGAN generator (test) | 22 | Apr 21, 2026 | ||
| Fine-Grained Visual Categorization | Flowers-102 | 22 | Feb 26, 2026 |