Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 1549 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Multi-hop Question Answering | G-Novel | 20 | Jun 2, 2026 | ||
| Multi-hop Question Answering | G-Medical | 20 | Jun 2, 2026 | ||
| Image Unlearning | Pins Face Recognition Random scenario | 20 | Jun 2, 2026 | ||
| Medical Image Segmentation | BUSI, ISIC-2017, ISIC-2018, Kvasir-SEG, GLAS | 20 | Jun 2, 2026 | ||
| Sparse Attack Detection | Gaussian noise with specific trajectory shift of 1.0 x 10^4 (m=150) (test) | 20 | Jun 2, 2026 | ||
| Backdoor Attack | Fear Trigger Emotion (OOD) | 20 | Jun 2, 2026 | ||
| Backdoor Attack | Fear Trigger Emotion (Generalization) | 20 | Jun 2, 2026 | ||
| Backdoor Attack | Fear Trigger Emotion (Standard) | 20 | Jun 2, 2026 | ||
| Image Compression | Kodak | 20 | Jun 23, 2026 | ||
| Image Classification | ImageNet-1K Clean | 20 | Jun 23, 2026 | ||
| Code Generation | CodeX | 20 | Jun 1, 2026 | ||
| Truthfulness | TruthfulQA | 20 | Jun 1, 2026 | ||
| Time-series forecasting | ILI standard (test) | 20 | Jun 1, 2026 | ||
| Question Answering | TruthfulQA | 20 | Jun 1, 2026 | ||
| End-to-end attention latency measurement | Apple M2 | 20 | Jun 1, 2026 | ||
| End-to-end attention latency measurement | RK3588S2 | 20 | Jun 1, 2026 | ||
| Reading Comprehension | DROP | 20 | Jun 1, 2026 | ||
| Question Answering | BoolQ | 20 | Jun 1, 2026 | ||
| Coreference Resolution | WSC | 20 | Jun 1, 2026 | ||
| Question Answering | SQuAD | 20 | Jun 1, 2026 | ||
| Natural Language Inference | CB | 20 | Jun 1, 2026 | ||
| Ad-Injected Response Generation | CA-Prod | 20 | Jun 1, 2026 | ||
| Probability Forecasting | Polymarket 30% held-out evaluation cliques | 20 | May 29, 2026 | ||
| Novel View Synthesis | RGBT-Scenes (test) | 20 | May 29, 2026 | ||
| Machine Translation | WMT Xx ⇒ En 24++ | 20 | May 29, 2026 |