Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 1042 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Multi-organ Nucleus Segmentation | MoNuSeg (test) | 27 | Feb 26, 2026 | ||
| Chest X-ray classification | CheXpert (test) | 27 | Apr 28, 2026 | ||
| Financial Question Answering | FinanceIQ | 27 | Feb 26, 2026 | ||
| Co-speech 3D Gesture Synthesis | BEAT2 (test) | 27 | Feb 26, 2026 | ||
| Place Recognition | HeLiPR Single-Session KAIST05 DB → KAIST05 Query 1.0 | 26 | Jul 9, 2026 | ||
| Long-form Question Answering | AlpacaFact FullP (100% polluted) | 26 | Jul 7, 2026 | ||
| Long-form Question Answering | AlpacaFact MixP (50% polluted) | 26 | Jul 7, 2026 | ||
| Long-form Question Answering | Biography FullP (100% polluted) | 26 | Jul 7, 2026 | ||
| Long-form Question Answering | Biography MixP (50% polluted) | 26 | Jul 7, 2026 | ||
| Long-form Question Answering | LongFact FullP (100% polluted) | 26 | Jul 7, 2026 | ||
| Long-form Question Answering | LongFact MixP (50% polluted) | 26 | Jul 7, 2026 | ||
| Long-form Question Answering | FAVA FullP (100% polluted) | 26 | Jul 7, 2026 | ||
| Long-form Question Answering | FAVA MixP (50% polluted) | 26 | Jul 7, 2026 | ||
| Mathematical Reasoning | AIME and HMMT Average | 26 | Jul 7, 2026 | ||
| Mathematical Reasoning | HMMT 2511 | 26 | Jul 7, 2026 | ||
| Mathematical Reasoning | HMMT 2502 | 26 | Jul 7, 2026 | ||
| System Verification and False Completion Analysis | Synthetic tasks 5 families, 50 tasks Section 9.2 generator | 26 | Jul 7, 2026 | ||
| AI-generated image detection | ConImageGen | 26 | Jul 7, 2026 | ||
| Malicious prompt detection | VLGuard & MSSBench π = 0.005 | 26 | Jul 7, 2026 | ||
| Malicious prompt detection | VLGuard & MLLMGuard π = 0.005 | 26 | Jul 7, 2026 | ||
| Malicious prompt detection | JailBreakV & GPT4V π = 0.005 | 26 | Jul 7, 2026 | ||
| Text Classification | TREC | 26 | Jul 3, 2026 | ||
| Trigger inversion | Yahoo (test) | 26 | Jul 1, 2026 | ||
| Trigger inversion | Yahoo high poison rate | 26 | Jul 1, 2026 | ||
| Image Classification | GTSRB | 26 | Jul 3, 2026 |