Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 510 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Egocentric Video Understanding | EgoSchema | 42 | Jun 16, 2026 | ||
| 3D Medical Image Segmentation | LA dataset | 42 | Mar 25, 2026 | ||
| Blind Image Quality Assessment | KonIQ | 42 | Jun 30, 2026 | ||
| Vulnerability Detection | CVE-Solidity (test) | 42 | Feb 26, 2026 | ||
| Bug Reachability | DS1 (custom oracles) | 42 | Feb 26, 2026 | ||
| IPI Detection | FIPI (test) | 42 | Feb 26, 2026 | ||
| Indirect Prompt Injection | HotpotQA | 42 | Feb 26, 2026 | ||
| Indirect Prompt Injection | Multi-News | 42 | Feb 26, 2026 | ||
| Text Generation | Medical Chatbot | 42 | Feb 26, 2026 | ||
| Language Modeling | MMLU | 42 | Apr 23, 2026 | ||
| Membership Inference Attack | GTSRB (test) | 42 | Feb 26, 2026 | ||
| Multi-hop QA | 2Wiki | 42 | May 25, 2026 | ||
| Multi-hop QA Retrieval | MuSiQue (test) | 42 | Jun 16, 2026 | ||
| Hallucination Detection | CoQA | 42 | Feb 26, 2026 | ||
| GUI Agent Task Completion | OSWorld 1.0 (test) | 42 | Jul 1, 2026 | ||
| Question Answering | AbsR | 42 | Feb 26, 2026 | ||
| Creative Writing | WritingBench | 42 | Jun 25, 2026 | ||
| Spam Detection | SPAMASSASSIN SPAM | 42 | Feb 26, 2026 | ||
| Transliteration | Aksharantar Indic-to-Roman v1 (test) | 42 | Feb 26, 2026 | ||
| Transliteration | Aksharantar Roman-to-Indic v1 (test) | 42 | Feb 26, 2026 | ||
| Value Alignment | ValueKaleidoscope VITAL Steerable setting | 42 | Feb 26, 2026 | ||
| Steerable Value Alignment | OpinionQA | 42 | Feb 26, 2026 | ||
| Value Alignment | VITAL Distributional MORALCHOICE | 42 | Feb 26, 2026 | ||
| Value Alignment | GLOBALOPINIONQA VITAL Distributional 6 values | 42 | Feb 26, 2026 | ||
| Logic Reasoning | Zebralogic | 42 | Feb 26, 2026 |