Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 1898 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Node Classification | pubmed 10%/10%/80% split | 18 | Feb 26, 2026 | ||
| Node Classification | facebook (10%/10%/80% split) | 18 | Feb 26, 2026 | ||
| Node Classification | amazon-ratings 10%/10%/80% | 18 | Feb 26, 2026 | ||
| Retrieval | SCIDOCS | 18 | Mar 10, 2026 | ||
| Reasoning | Downstream Reasoning Benchmarks (MATH, GSM8K, AQUA, AIME, AMC, MMLU, GPQA) | 18 | Feb 26, 2026 | ||
| Code generation | HumanEval | 18 | Mar 5, 2026 | ||
| Graph Classification | molbbbp | 18 | Apr 20, 2026 | ||
| WebShell Detection | 26.59K PHP scripts dataset (test) | 18 | Feb 26, 2026 | ||
| Graph Structure Learning | Autoregressive (AR) model p=120 synthetic (test) | 18 | Feb 26, 2026 | ||
| Graph Structure Learning | Moving Average (MA) model (p=120) synthetic (test) | 18 | Feb 26, 2026 | ||
| Multiclass classification | acoustic (test) | 18 | Feb 26, 2026 | ||
| Multiclass classification | letter (test) | 18 | Feb 26, 2026 | ||
| Causal Discovery | SF5 (n=30, h=5) | 18 | Feb 26, 2026 | ||
| Causal Discovery | ER5 (n=30, h=5) | 18 | Feb 26, 2026 | ||
| Toy target distribution sampling | GMM40 d = 50 | 18 | May 20, 2026 | ||
| Variational Inference | State-Space model (train) | 18 | Feb 26, 2026 | ||
| Classification | Gaussian mixture synthetic dataset target domain (test) | 18 | Feb 27, 2026 | ||
| Single-turn dialogue | Anthropic HH | 18 | May 4, 2026 | ||
| Clustering | Synthetic Dataset Setup 1.0 (test) | 18 | Feb 26, 2026 | ||
| Pressure Synthesis | SLP (test) | 18 | Feb 26, 2026 | ||
| Cortical layer segmentation | macaque V1 | 18 | Feb 26, 2026 | ||
| Disease Classification | Chinese (external val) | 18 | Feb 26, 2026 | ||
| Disease Classification | Stanford external (val) | 18 | Feb 26, 2026 | ||
| Disease Classification | Johns Hopkins (JHU) (internal) | 18 | Feb 26, 2026 | ||
| Point Cloud Lossless Attribute Compression | MPEG CAT1 Sparse | 18 | Feb 26, 2026 |