Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 553 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Calibration | Brock-Hommes (test) | 40 | Feb 26, 2026 | ||
| Parameter Calibration | Brock–Hommes problems (various parameter sets) | 40 | Feb 26, 2026 | ||
| Parameter estimation | Brock–Hommes problems (test) | 40 | Feb 26, 2026 | ||
| Program synthesis | HumanEval (test) | 40 | Jun 4, 2026 | ||
| Classification | Synthetic Dynamical Systems Lorenz, Thomas, and Hindmarsh-Rose (test) | 40 | Feb 26, 2026 | ||
| Judge Alignment | WMT ZH-EN | 40 | Feb 26, 2026 | ||
| Concept Attribution | Sarcasm Dataset | 40 | Feb 26, 2026 | ||
| Unlearning | Syntax-preserving dataset (forget set) | 40 | Feb 26, 2026 | ||
| Language Modeling | C4 LLaMA-130M (val) | 40 | May 14, 2026 | ||
| Node Classification | German | 40 | Jun 4, 2026 | ||
| Node Classification | PubMed 10-shot | 40 | Jun 11, 2026 | ||
| Node Classification | CiteSeer 10-shot | 40 | Jun 11, 2026 | ||
| Node Classification | Cora 10-shot | 40 | Jun 11, 2026 | ||
| Unconditional 3D Point Cloud Generation | ShapeNet airplane (val) | 40 | Feb 26, 2026 | ||
| Solving Hamilton-Jacobi equations | Hamilton-Jacobi equation with quadratic initial condition and Hamiltonian | 40 | Feb 26, 2026 | ||
| Image Classification | CIFAR100 (test) | 40 | Feb 26, 2026 | ||
| Image Classification | CIFAR10 (test) | 40 | Feb 26, 2026 | ||
| Unconditional generation | OpenWebText L=1024 (test) | 40 | May 11, 2026 | ||
| Image Classification | CIFAR-10 standard (test) | 40 | Feb 26, 2026 | ||
| Probabilistic forecasting | ECL | 40 | Feb 26, 2026 | ||
| Probabilistic forecasting | Traffic | 40 | Feb 26, 2026 | ||
| Probabilistic forecasting | ETTh1 | 40 | Feb 26, 2026 | ||
| Probabilistic forecasting | ETTm1 | 40 | Feb 26, 2026 | ||
| Node Classification | Ogbn-arxiv | 40 | May 14, 2026 | ||
| Safety Refusal | Safety Evaluation Prompts | 40 | Feb 26, 2026 |