Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 326 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Image Retrieval | Oxford105k (test) | 56 | Feb 26, 2026 | ||
| Semantic Relatedness | SICK 2014 (test) | 56 | Feb 27, 2026 | ||
| Hypothesis Testing | 3-dimensional fractional Brownian motion (fBM) | 56 | Feb 26, 2026 | ||
| Image Generation | LSUN bedroom | 56 | Feb 26, 2026 | ||
| Few-shot Image Classification | mini-ImageNet K=20 (test) | 56 | Feb 26, 2026 | ||
| Linguistic Acceptability | CoLA (test) | 56 | Jun 15, 2026 | ||
| Machine Translation | WMT en-fr 14 | 56 | Feb 26, 2026 | ||
| Semantic Segmentation | NYU Depth V2 | 56 | Jun 4, 2026 | ||
| Uncertainty Estimation | SciQA | 56 | Apr 2, 2026 | ||
| 3D Point Cloud Generation | ShapeNet Chair category (test) | 56 | Feb 26, 2026 | ||
| Jailbreak Safety Evaluation | MM-Safety Bench (test) | 56 | Feb 26, 2026 | ||
| Face Identification | CASIA-WebFace (25 unseen identities) | 55 | Jul 8, 2026 | ||
| Instruction Following | ALFWorld (test) | 55 | Jul 7, 2026 | ||
| Image Classification | ImageNet Tiny | 55 | Jul 7, 2026 | ||
| Robotic Manipulation | libero goal | 55 | Jun 16, 2026 | ||
| Classification | UKB | 55 | Jun 30, 2026 | ||
| Open-ended hallucination | Open-ended hallucination benchmark | 55 | Jun 9, 2026 | ||
| Open-ended refusal | Open-ended refusal benchmark | 55 | Jun 9, 2026 | ||
| Refusal | Refusal benchmark | 55 | Jun 9, 2026 | ||
| Clustering | HandWritten V=6, K=10 | 55 | Jun 4, 2026 | ||
| Heart rate estimation | MCD N=200 FULLHDWEBCAM (test) | 55 | Jun 4, 2026 | ||
| Mislabeled data detection | PANDA | 55 | Jun 1, 2026 | ||
| Mislabeled data detection | DeepDRiD | 55 | Jun 1, 2026 | ||
| Mislabeled data detection | ISIC | 55 | Jun 1, 2026 | ||
| Predicting code correctness | LiveSQLBench SQLite | 55 | May 28, 2026 |