Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 1194 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Multi-task aerial image reasoning | HRVQA-VL (test) | 24 | May 7, 2026 | ||
| Multi-objective Optimization | DTLZ1-4 End Generation 1.0 (test) | 24 | May 7, 2026 | ||
| Multi-objective Optimization | DTLZ1-4 1.0 (intermediate generation) | 24 | May 7, 2026 | ||
| Multi-Source Domain Adaptation | Office-31 | 24 | May 7, 2026 | ||
| Multi-Source Domain Adaptation | ImageCLEF | 24 | May 7, 2026 | ||
| Language Model Evaluation | Qwen3-0.6B Evaluation Suite average | 24 | May 7, 2026 | ||
| Multimodal Language Understanding | Qwen3 1.7B | 24 | May 7, 2026 | ||
| Commonsense Reasoning | Winogrande | 24 | May 7, 2026 | ||
| Audio-Visual Classification | CREMA-D | 24 | Jul 9, 2026 | ||
| Long-tailed Class-Incremental Learning | ImageNet Subset-LT | 24 | May 6, 2026 | ||
| Long-tailed Class-Incremental Learning | CIFAR-100-LT | 24 | May 6, 2026 | ||
| Image Classification | ImageNet Subset (From Half) LT (test) | 24 | May 6, 2026 | ||
| Image Classification | CIFAR-100 From Half LT (test) | 24 | May 6, 2026 | ||
| Agent Safety Judgment | ROME Shortcut Decision-Making 1.0 | 24 | May 6, 2026 | ||
| Agent Safety Judgment | ROME Contextual Ambiguity 1.0 | 24 | May 6, 2026 | ||
| Agent Safety Judgment | ROME Implicit Risks 1.0 (IR) | 24 | May 6, 2026 | ||
| Agent Safety Judgment | ROME Original 1.0 (Seed) | 24 | May 6, 2026 | ||
| Node Classification | Cora | 24 | May 6, 2026 | ||
| GUI grounding | OSWorld-G-Refine | 24 | Jul 1, 2026 | ||
| Reconstruction of Discrete Branching Dynamics | CITE 50D (t=3) | 24 | May 4, 2026 | ||
| Reconstruction of discrete branching dynamics | 5D EB t=1 | 24 | May 4, 2026 | ||
| Trajectory interpolation | EB 50D (held-out time points) | 24 | May 4, 2026 | ||
| Table Reasoning | WTQ + FeTaQA + TabFact Combined (test) | 24 | May 4, 2026 | ||
| Hallucination detection | Summedits (test) | 24 | May 4, 2026 | ||
| Safety classification | BeaverTails (test) | 24 | May 4, 2026 |