Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 377 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Manipulation gain evaluation | Synthetic classification task | 50 | May 11, 2026 | ||
| Semantic Segmentation | SceneParse150 | 50 | May 19, 2026 | ||
| Image Classification | CIFAR-100 coarse (test) | 50 | May 11, 2026 | ||
| Classification | MNIST-SVHN r=50 | 50 | May 12, 2026 | ||
| Auto-bidding | AuctionNet Low reward sparsity | 50 | May 8, 2026 | ||
| Auto-bidding | AuctionNet Medium reward sparsity | 50 | May 8, 2026 | ||
| Auto-bidding | AuctionNet High reward sparsity | 50 | May 8, 2026 | ||
| Jailbreak | HarmBench | 50 | May 20, 2026 | ||
| Node Classification | CHAMELEON vanilla (test) | 50 | May 8, 2026 | ||
| Safety Evaluation | AdvBench | 50 | May 7, 2026 | ||
| Event Type Classification | TUEV | 50 | May 19, 2026 | ||
| Image Classification | ImageNet | 50 | May 14, 2026 | ||
| Math Problem Solving | OlympiadBench | 50 | Jun 23, 2026 | ||
| Multi-hop Question Answering | Average (MuSiQue, HotpotQA, 2WikiMultiHopQA, LongSeal) | 50 | Apr 28, 2026 | ||
| Multi-hop Question Answering | LongSeal | 50 | Apr 28, 2026 | ||
| Multi-hop Question Answering | HotpotQA | 50 | Apr 28, 2026 | ||
| Multi-hop Question Answering | MuSiQue | 50 | Apr 28, 2026 | ||
| Binary safety hallucination detection | EndoVis18-VQA Out-of-template (val) | 50 | Apr 24, 2026 | ||
| Backdoor Detection | LMSYS-Chat-1M | 50 | Apr 23, 2026 | ||
| Backdoor Detection | MMLU | 50 | Apr 23, 2026 | ||
| Backdoor Detection | Calibration Set | 50 | Apr 23, 2026 | ||
| Backdoor Detection | Poisoned Prompt Set | 50 | Apr 23, 2026 | ||
| Chemical Property Prediction | Polymers (5-fold cross-val) | 50 | Apr 20, 2026 | ||
| Packing | PWT Multiple Instances Deterministic | 50 | Apr 16, 2026 | ||
| Multi-Hop Question Answering | Bamboogle (test) | 50 | Jun 12, 2026 |