Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 2170 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Visual Question Answering | Text VQA | 16 | May 12, 2026 | ||
| Jailbreak Defense | Prefix-Probe-600 Vicuna-7B | 16 | May 12, 2026 | ||
| Clutter removal | Pile scenes single-view, fixed camera, gamma noise | 16 | May 12, 2026 | ||
| Clutter removal | Packed scenes single-view, fixed camera, gamma noise | 16 | May 12, 2026 | ||
| Multi-Robot Motion Planning | Unicycle (UC) Swap (N=4, 5, 8, 10, 15, 20, 25, 30) | 16 | May 12, 2026 | ||
| Science Question Answering | DeepSeek-distilled ARC-C (test) | 16 | May 12, 2026 | ||
| Mathematical Reasoning | GSM8K DeepSeek-distilled (test) | 16 | May 12, 2026 | ||
| Malware Detection | EMBER Domain | 16 | May 12, 2026 | ||
| Digital Watermarking | CelebA | 16 | May 12, 2026 | ||
| Digital Watermarking | COCO | 16 | May 12, 2026 | ||
| Object Detection | COCO 2017 | 16 | May 12, 2026 | ||
| Natural Language Understanding | PIQA | 16 | May 21, 2026 | ||
| Dose Prediction | GDP-HMM (test) | 16 | Jun 29, 2026 | ||
| Question Answering | MMLU-Pro (full) | 16 | May 12, 2026 | ||
| Image Classification | ImageNet base-to-novel | 16 | May 18, 2026 | ||
| Simple P-log Program Computation | robot | 16 | May 12, 2026 | ||
| Sparse Probing | SAEBench | 16 | May 12, 2026 | ||
| Reconstruction | SAEBench held-out data | 16 | May 12, 2026 | ||
| Utility Preservation | MMLU 5-shot (test) | 16 | May 12, 2026 | ||
| Imprint Forgery Attack | ImageNet | 16 | May 12, 2026 | ||
| Imprint Removal Attack | ImageNet (test) | 16 | May 12, 2026 | ||
| Imprint Forgery Attack | MS-COCO (test) | 16 | May 12, 2026 | ||
| Question Answering | HotpotQA | 16 | May 12, 2026 | ||
| Sequential Model Editing | ZsRE T = 5000 | 16 | May 12, 2026 | ||
| Jailbreak attack evaluation | 500 randomly sampled prompts (test) | 16 | May 12, 2026 |