Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 1356 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Hidden Objective Detection | Hidden Objective Detection Evaluation Set Monitor Notice 1.0 (out-of-domain) | 22 | May 14, 2026 | ||
| Instruction-based Image Editing | GEdit-Bench-EN (test) | 22 | May 13, 2026 | ||
| Reasoning | GPQA D | 22 | May 13, 2026 | ||
| Question Answering | CM17k | 22 | May 13, 2026 | ||
| Cross-view Retrieval | University-1652 Satellite -> Drone | 22 | May 13, 2026 | ||
| Out-of-Domain Reasoning Aggregation | OOD Average | 22 | May 13, 2026 | ||
| Logical Reasoning | LSAT-AR | 22 | May 13, 2026 | ||
| Multi-step Narrative Reasoning | MUSR | 22 | May 13, 2026 | ||
| Scientific Question Answering | GPQA Diamond | 22 | May 13, 2026 | ||
| CTR prediction | 4-Market (Overall) | 22 | May 13, 2026 | ||
| CTR prediction | 3-Market (Overall) | 22 | May 13, 2026 | ||
| CTR prediction | 2-Market (Overall) | 22 | May 13, 2026 | ||
| Visual Perception | CVBench 2D | 22 | Jul 7, 2026 | ||
| Motor Imagery Classification | SHU-MI 2-Class | 22 | Jul 7, 2026 | ||
| Video Question Answering | MVBench, VideoMME, LVB, MLVU v1 (test val) | 22 | May 13, 2026 | ||
| Vision-Language Question Answering | Pooled Shared (GQA, POPE, LLaVA-Wild, MMMU Pro, GMAI-MMBench, MME-Finance) (test) | 22 | May 12, 2026 | ||
| Novel view interpolation | Dynamic Indoor Scene Dataset | 22 | May 12, 2026 | ||
| Future frame extrapolation | Dynamic Object Dataset | 22 | May 12, 2026 | ||
| Spatial-Temporal Super-Resolution | Polarization Video Time ×2 | 22 | May 12, 2026 | ||
| Visual Grounding | MMB v1.1 | 22 | May 12, 2026 | ||
| Visual Grounding | GQA | 22 | May 12, 2026 | ||
| Visual Grounding | MMStar | 22 | May 12, 2026 | ||
| Visual Grounding | RWQA | 22 | May 12, 2026 | ||
| Optic Cup Segmentation | RIGA+ Messidor Target Domains (test) | 22 | May 12, 2026 | ||
| Optic Disc Segmentation | RIGA+ Messidor Target Domains (test) | 22 | May 12, 2026 |