Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 935 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Scenario Reduction | SEL Large-scale 50 instances (test) | 28 | May 15, 2026 | ||
| Scenario Reduction | SEL Medium-scale 50 instances (test) | 28 | May 15, 2026 | ||
| Multi-turn Agentic Task | PDDL | 28 | May 19, 2026 | ||
| Multi-turn Agentic Task | ScienceWorld | 28 | May 19, 2026 | ||
| Pixel-level forgery localization | DocTamperFCD, DocTamperSCD, DocTamperTest, T-SROIE, Tampered IC13, OSTF, RTM document-oriented (full) | 28 | May 15, 2026 | ||
| Machine Unlearning | TOFU (Split99) | 28 | May 15, 2026 | ||
| Chart Question Answering | ChartQA | 28 | Jun 17, 2026 | ||
| Language Modeling | C4 en (val) | 28 | Jul 7, 2026 | ||
| Air Quality Estimation | China air quality data meta (test) | 28 | May 14, 2026 | ||
| Question Answering and Commonsense Reasoning | NLP Benchmark Suite Zero-shot (HellaSwag, RACE, PIQA, WinoGrande, ARC, OBQA) (test) | 28 | May 14, 2026 | ||
| Link Prediction | Bili_Dance | 28 | Jun 19, 2026 | ||
| Knowledge Graph Question Answering | WebQSP | 28 | May 13, 2026 | ||
| 1-year Pancreatic Cancer | EHRSHOT (test) | 28 | May 13, 2026 | ||
| 30-day Readmission | EHRSHOT (test) | 28 | May 13, 2026 | ||
| Logical Reasoning | KK | 28 | Jun 2, 2026 | ||
| Instruction Following | Alpaca poisoned (test) | 28 | May 13, 2026 | ||
| Classification | Shuttle | 28 | Jun 23, 2026 | ||
| Vulnerable Agent Identification | Taxi environment 100 agents, 50 adversaries | 28 | May 13, 2026 | ||
| Vulnerable Agent Identification | Taxi environment 50 agents, 25 adversaries | 28 | May 13, 2026 | ||
| Peak Estimation | Goodreads per-genre | 28 | May 13, 2026 | ||
| Code Generation | LiveCodeBench | 28 | Jun 16, 2026 | ||
| Bandit Optimization | Setting 4 | 28 | May 12, 2026 | ||
| Bandit Optimization | Setting 3 | 28 | May 12, 2026 | ||
| Stepwise tool-use | API-Bank (test) | 28 | May 12, 2026 | ||
| Planning Evaluation | NAVSIM navhard v2 | 28 | May 12, 2026 |