Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 1823 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Clinical time-series prediction | MIMIC-III (test) | 18 | May 21, 2026 | ||
| Web Agent | WorkArena L2 | 18 | May 21, 2026 | ||
| Web Agent | WorkArena L1 | 18 | May 21, 2026 | ||
| Web Agent | WebArena Lite | 18 | May 21, 2026 | ||
| Refusal Steering | SALAD Alpaca HarmBench | 18 | May 21, 2026 | ||
| Class-conditional image generation | truck concept | 18 | May 21, 2026 | ||
| Node Label Frequency Proportionality | CiteSeer | 18 | May 21, 2026 | ||
| Node Label Frequency Proportionality | 18 | May 21, 2026 | |||
| Interpretive Quality Evaluation | Interpretive Quality Evaluation Dataset | 18 | May 21, 2026 | ||
| Evidence-Anchored Consolidation | Evidence-Anchored Consolidation Evaluation | 18 | May 21, 2026 | ||
| Prompt-level detection | matched-PP benchmark α = 1 (5-fold stratified CV) | 18 | May 20, 2026 | ||
| Long-context Memory Question Answering | LoCoMo (test) | 18 | May 20, 2026 | ||
| Test-time adaptation | ITD γ = 10⁻¹ (test) | 18 | May 20, 2026 | ||
| Test-time adaptation | ImageNet-C γ = 10⁻¹ (test) | 18 | May 20, 2026 | ||
| Grounding | refCOCO (testB) | 18 | May 29, 2026 | ||
| Reflective Visual Generation | R3-Bench | 18 | May 20, 2026 | ||
| Long-context Reasoning | Long-context Reasoning Suite (test) | 18 | May 20, 2026 | ||
| Multimodal NP-hard combinatorial optimization | MM-NP-Bench | 18 | May 20, 2026 | ||
| Distractor Generation | CLOTH | 18 | May 20, 2026 | ||
| Multi-hop Question Answering | Bamboogle | 18 | May 20, 2026 | ||
| Multi-hop Question Answering | MuSiQue | 18 | May 20, 2026 | ||
| Multi-hop Question Answering | 2Wiki | 18 | May 20, 2026 | ||
| Question Answering | PopQA | 18 | May 20, 2026 | ||
| Question Answering | TriviaQA | 18 | May 20, 2026 | ||
| Question Answering | NQ | 18 | May 20, 2026 |