Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 951 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Intent Classification | FSC (Fluent Speech Commands) | 28 | Jun 26, 2026 | ||
| Language Modeling | Wikitext-103 | 28 | May 28, 2026 | ||
| Multi-robot exploration | 80x80 world, 16 Agents, 30% static obstacle density 100 runs | 28 | Mar 10, 2026 | ||
| Multi-robot exploration | 40x40 world, 4 Agents, 30% static obstacle density (100 runs) | 28 | Mar 10, 2026 | ||
| Image Classification | Flowers, CUB, Cars, CIFAR-10, CIFAR-100, Food, iNat downstream evaluation | 28 | Mar 10, 2026 | ||
| Desnowing | SRRS | 28 | Jun 23, 2026 | ||
| Node Classification | CiteSeer | 28 | Mar 10, 2026 | ||
| Verification of Digital Agent Trajectories | VisualWebArena (VWA) and OSWorld Trajectories | 28 | Mar 10, 2026 | ||
| Task-Incremental Learning | CIFAR-100 Seq | 28 | Apr 23, 2026 | ||
| Task-Incremental Learning | Seq-CIFAR-10 | 28 | Mar 10, 2026 | ||
| Binary Classification | sonar | 28 | Apr 10, 2026 | ||
| Binary Classification | plrx | 28 | Apr 10, 2026 | ||
| Image-Text Retrieval | MS-COCO 5K (test) | 28 | Apr 21, 2026 | ||
| Multi-label class-incremental learning | MS-COCO 2014 (B0-C10) | 28 | Mar 9, 2026 | ||
| Mathematical Reasoning | MSVAMP | 28 | Jun 18, 2026 | ||
| Molecule generation | GuacaMol | 28 | Apr 21, 2026 | ||
| Node Classification | Pubmed | 28 | Mar 6, 2026 | ||
| Node Classification | Cora | 28 | Mar 6, 2026 | ||
| Graph Backdoor Attack | Arxiv | 28 | Mar 6, 2026 | ||
| Graph Backdoor Attack | Flickr | 28 | Mar 6, 2026 | ||
| Graph Backdoor Attack | Pubmed | 28 | Mar 6, 2026 | ||
| Graph Backdoor Attack | Cora | 28 | Mar 6, 2026 | ||
| Criminal Sentencing, Legal Validity, and Suspect-Level Performance Evaluation | LEEC | 28 | May 27, 2026 | ||
| Long Video Understanding | Long Video Bench | 28 | Mar 6, 2026 | ||
| Long Video Understanding | MLVU v1.0 (test) | 28 | Apr 21, 2026 |