Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 538 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Sycophancy Mitigation | Held-out | 40 | Jun 5, 2026 | ||
| Sycophancy Mitigation | MMLU on-the-fly | 40 | Jun 5, 2026 | ||
| Deepfake detection | DeepfakeTIMIT Dataset2 (K-Fold Cross Validation) | 40 | Jun 5, 2026 | ||
| Image Deblurring | Standard Images Average blur kernel (test) | 40 | Jun 5, 2026 | ||
| Image Deblurring | Standard Test Images Gaussian blur kernel | 40 | Jun 5, 2026 | ||
| Adversarial Robustness | CIFAR-10 | 40 | Jun 4, 2026 | ||
| Text-guided image editing | PIE-Bench | 40 | Jun 17, 2026 | ||
| Strategy Games | LICHESS | 40 | Jun 2, 2026 | ||
| Code Generation | CodeProjectEval (Per-project) | 40 | Jun 2, 2026 | ||
| Trajectory Prediction | SHRP2 | 40 | Jun 2, 2026 | ||
| Trajectory Prediction | highD | 40 | Jun 2, 2026 | ||
| Active Learning | GC | 40 | Jun 9, 2026 | ||
| Active Learning | AC | 40 | Jun 9, 2026 | ||
| Active Learning | IvsJ | 40 | Jun 9, 2026 | ||
| Grammar-Constrained Decoding | SMILES | 40 | Jun 2, 2026 | ||
| Grammar-Constrained Decoding | JSON | 40 | Jun 2, 2026 | ||
| Grammar-Constrained Decoding | C++ | 40 | Jun 2, 2026 | ||
| Regression | ZIP Code weighted avg. 29 tasks (cross-regional) | 40 | Jun 2, 2026 | ||
| Capability Self-Assessment | MMLU-Pro Science | 40 | Jun 2, 2026 | ||
| Capability Self-Assessment | Math | 40 | Jun 2, 2026 | ||
| Retrieval | Victorian Crime Diaries (VCD) | 40 | Jun 2, 2026 | ||
| Named Entity Recognition | ener | 40 | Jun 2, 2026 | ||
| Named Entity Recognition | wnut | 40 | Jun 2, 2026 | ||
| Solving Helmholtz Equation | Helmholtz Equation | 40 | May 29, 2026 | ||
| Solving Nonlinear Burger's Equation | Nonlinear Burger's equation ζ = 1 | 40 | May 29, 2026 |