Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 1890 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Text-to-Image Safety | UD | 18 | Feb 26, 2026 | ||
| Text-to-Image Safety | T2VSafetyBench | 18 | Feb 26, 2026 | ||
| Indirect Prompt Injection Defense | TrojanTools | 18 | Feb 26, 2026 | ||
| Indirect Prompt Injection Defense | Combined Attacks | 18 | Feb 26, 2026 | ||
| Indirect Prompt Injection Defense | Ignore Instruction | 18 | Feb 26, 2026 | ||
| Watermark Decoding | COCO (subset) | 18 | Feb 26, 2026 | ||
| Image Classification | 11 datasets average CLIP ResNet-101 (test) | 18 | Feb 26, 2026 | ||
| Attention heatmap prediction | SurgAtt-SZPH (test) | 18 | Feb 26, 2026 | ||
| Math reasoning | MATH-500 (in-distribution) | 18 | May 21, 2026 | ||
| End-to-End Inference | LMSys-Chat-1M and ShareGPT traces | 18 | Feb 26, 2026 | ||
| Autonomous Driving | NAVSIM navtest v2 (test) | 18 | Mar 17, 2026 | ||
| 3D Object Detection | K-Radar (test) | 18 | Jul 3, 2026 | ||
| Retrieval | Clipper | 18 | Feb 27, 2026 | ||
| General Question Answering | PopQA | 18 | Feb 26, 2026 | ||
| General Question Answering | TriviaQA | 18 | Feb 26, 2026 | ||
| Multi-image context generation | MICON-Bench | 18 | Feb 26, 2026 | ||
| Speech Recognition | LRS3 high-resource | 18 | Feb 26, 2026 | ||
| Speech Recognition | LRS3 low-resource | 18 | Feb 26, 2026 | ||
| Binary Classification | electricity | 18 | May 14, 2026 | ||
| Fault Diagnosis | DIRG (Total) | 18 | Feb 26, 2026 | ||
| Fault Diagnosis | DIRG (Defective) | 18 | Feb 26, 2026 | ||
| Fault Diagnosis | DIRG (Non-Defective) | 18 | Feb 26, 2026 | ||
| Fault Diagnosis | HIT Bearing Dataset (test) | 18 | Feb 26, 2026 | ||
| Face Anti-Spoofing | WMCA | 18 | Mar 4, 2026 | ||
| Face Anti-Spoofing | SIW-M V2 | 18 | Mar 4, 2026 |