Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 649 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Document Retrieval | MS MARCO MS300K (test) | 36 | May 26, 2026 | ||
| Sequential Recommendation | Amazon Reviews 8 Domains | 36 | Feb 27, 2026 | ||
| Sequential Recommendation | Amazon (test) | 36 | Feb 27, 2026 | ||
| Reciprocal Recommendation | Synthetic datasets n=75, m=50, e=log 10 independent trials | 36 | Feb 26, 2026 | ||
| Sequential Recommendation | Toys (test) | 36 | Mar 5, 2026 | ||
| Next-POI Recommendation | NYC (test) | 36 | May 13, 2026 | ||
| Spatial Reasoning | RoboSpatial | 36 | Mar 24, 2026 | ||
| Long Video Understanding | MLVU 3-120 min | 36 | Apr 21, 2026 | ||
| Real-world Super-Resolution | DIV2K real-world degradation (test) | 36 | Feb 26, 2026 | ||
| Text-to-Image Generation | DPG-Bench | 36 | Mar 27, 2026 | ||
| All-in-One Image Restoration | CBSD68, SOTS, and Rain100L | 36 | May 12, 2026 | ||
| Window-level anti-pattern detection | TorchTraceAP Seen | 36 | Feb 26, 2026 | ||
| Uncertainty Quantification | Vision Datasets averaged (test) | 36 | Feb 26, 2026 | ||
| Face Template Inversion Attack | LFW | 36 | May 4, 2026 | ||
| 3D Reconstruction | DDAD | 36 | Jul 8, 2026 | ||
| Music-to-Dance Generation | FineDance | 36 | May 28, 2026 | ||
| Object Shuffling Prediction | ObjShuf | 36 | Feb 26, 2026 | ||
| Object Existence Prediction | ObjExist | 36 | Feb 26, 2026 | ||
| Multimodal Question Answering | MMQA | 36 | Feb 26, 2026 | ||
| Moving Direction Prediction | MoveDir | 36 | Feb 26, 2026 | ||
| Image Editing | IEdit | 36 | Feb 26, 2026 | ||
| Counterfactual Inference | CFInfer | 36 | Feb 26, 2026 | ||
| Action Prediction | ActPred | 36 | Feb 26, 2026 | ||
| Quality Assessment | AIGCIQA 2023 | 36 | Feb 26, 2026 | ||
| Segmentation | BraTS 2020 | 36 | Feb 26, 2026 |