Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 1232 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Mathematical Reasoning | AIME 2024 | 24 | Feb 26, 2026 | ||
| Mathematical Reasoning | MATH-500 | 24 | Feb 26, 2026 | ||
| Mathematical Reasoning | GSM8K | 24 | Feb 26, 2026 | ||
| Document Reranking | BEIR | 24 | Apr 13, 2026 | ||
| Multi-label Abnormality Analysis | CT-RATE (test) | 24 | Jun 1, 2026 | ||
| Adversarial Purification | CIFAR-10 (test) | 24 | Feb 26, 2026 | ||
| General Language Modeling | MMLU, ARC-Challenge, and CommonsenseQA Aggregate | 24 | Feb 26, 2026 | ||
| Vulnerability Detection | PrimeVul Seed: PrimeVul Train (test) | 24 | Feb 26, 2026 | ||
| Vulnerability Detection | BigVul Seed: Devign (test) | 24 | Feb 26, 2026 | ||
| Image Classification | CIFAR-10 | 24 | Feb 26, 2026 | ||
| Mathematical Reasoning | AMC23 (val) | 24 | Apr 14, 2026 | ||
| Data Assimilation | SQG | 24 | Feb 26, 2026 | ||
| Time-series interval forecasting | Electricity Demand | 24 | Feb 26, 2026 | ||
| Time-series interval forecasting | Google Stock | 24 | Feb 26, 2026 | ||
| Time-series interval forecasting | Amazon Stock | 24 | Feb 26, 2026 | ||
| Post-shift coverage recovery | Changepoint Shift 2 (Time 2) | 24 | Feb 26, 2026 | ||
| Post-shift coverage recovery | Changepoint Shift 1 (Time 1) | 24 | Feb 26, 2026 | ||
| Online Conformal Prediction | Heavy-tailed | 24 | Feb 26, 2026 | ||
| Online Conformal Prediction | Variance Changepoint | 24 | Feb 26, 2026 | ||
| Conformal Prediction | Distribution Drift simulated (test) | 24 | Feb 26, 2026 | ||
| Conformal Prediction | Changepoint simulated (test) | 24 | Feb 26, 2026 | ||
| Online Conformal Prediction | Distribution Drift | 24 | Feb 26, 2026 | ||
| Average Treatment Effect (ATE) Estimation | Synthetic Heterogeneous Treatment Effects (HTE) Data (test) | 24 | Feb 26, 2026 | ||
| Image Generation | Smithsonian Butterflies (test) | 24 | Feb 26, 2026 | ||
| Missing Data Imputation | eICU Collaborative Research Database Simulation of Blockwise Missing Data n=5000, p=40 | 24 | Feb 26, 2026 |