Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 2040 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| 4D Spatial-Temporal Reasoning | MLLM4D-Bench | 17 | Mar 4, 2026 | ||
| Long-Video Understanding | LongVideoBench 8s~60m | 17 | Mar 4, 2026 | ||
| Affect recognition | WESAD (test) | 17 | Mar 13, 2026 | ||
| Safety Evaluation | Aggregate JBB, SR, WJ | 17 | Mar 4, 2026 | ||
| Safety Evaluation | StrongReject (SR) | 17 | Mar 4, 2026 | ||
| Jailbreak Safety | WildJailbreak | 17 | Mar 4, 2026 | ||
| Jailbreak Safety | JailbreakBench | 17 | Mar 4, 2026 | ||
| Video Object Segmentation | MOSE v2 (val) | 17 | Mar 31, 2026 | ||
| Video Object Segmentation | MOSE v1 (val) | 17 | Mar 31, 2026 | ||
| Multi-hop Question Answering | 2Wiki | 17 | Mar 4, 2026 | ||
| Multi-hop Question Answering | HotPotQA | 17 | Mar 4, 2026 | ||
| Zero-shot Question Answering | Benchmarks (ArcC, ArcE, PiQA, Wino) Zero-shot | 17 | Mar 4, 2026 | ||
| Multimodal Mathematical Reasoning | MathVista | 17 | Mar 4, 2026 | ||
| Image Classification | ImageNet HR | 17 | Mar 4, 2026 | ||
| Mathematical Reasoning | MathVista | 17 | Mar 4, 2026 | ||
| Zero-shot Prediction | ARC-C | 17 | Feb 27, 2026 | ||
| Response Selection | MWOZ 2.1 | 17 | Feb 27, 2026 | ||
| Multi-hop Question Answering | 2WikiMHQA (test) | 17 | Feb 27, 2026 | ||
| Multi-hop Question Answering | MuSiQue v1 (test) | 17 | Feb 27, 2026 | ||
| Multi-hop Question Answering | 2WikiMHQA in-distribution | 17 | Feb 27, 2026 | ||
| Multi-hop Question Answering | HotpotQA in-distribution | 17 | Feb 27, 2026 | ||
| Multi-hop Question Answering | MuSiQue in-distribution | 17 | Feb 27, 2026 | ||
| Music-to-Dance | AIST++ | 17 | Feb 27, 2026 | ||
| Vision-Language Evaluation | MME (test) | 17 | Feb 27, 2026 | ||
| Visual Captioning | MS-COCO English In-domain (test) | 17 | Feb 27, 2026 |