Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 2652 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Multi-joint waypoint prediction in 3D space | H2O-PT | 14 | May 12, 2026 | ||
| Socratic Conversation Generation | Socratic Debugging of Student Code 227 samples | 14 | May 12, 2026 | ||
| Reasoning Trajectory Generation | Socratic Debugging of Student Code 227 samples | 14 | May 12, 2026 | ||
| Causal Discovery | NetSim Sim3 10 nodes 15 edges | 14 | May 12, 2026 | ||
| Logical Reasoning | SLR-BENCH | 14 | May 12, 2026 | ||
| Person Re-Identification | Market-MM | 14 | May 12, 2026 | ||
| Code generation | HumanEval | 14 | Jun 9, 2026 | ||
| Kinodynamic Multi-agent Path Planning | Narrow Corridor UC N=3 | 14 | May 12, 2026 | ||
| Mathematical Reasoning | GSM8K CoT | 14 | May 12, 2026 | ||
| Mathematical Reasoning | MATH | 14 | May 12, 2026 | ||
| Video Summarization | SumMe (5FCV) | 14 | May 12, 2026 | ||
| Motion Deblurring | CelebA 256 x 256 | 14 | May 12, 2026 | ||
| Gaussian Deblurring | CelebA 256 x 256 | 14 | May 12, 2026 | ||
| Inpainting | CelebA 256 x 256 | 14 | May 12, 2026 | ||
| Super-resolution | CelebA 256 x 256 | 14 | May 12, 2026 | ||
| Reading Comprehension | RACE | 14 | May 12, 2026 | ||
| Relation Extraction | RE | 14 | May 12, 2026 | ||
| Question Answering | TQA | 14 | May 12, 2026 | ||
| Reading Comprehension | DuoRC | 14 | May 12, 2026 | ||
| Question Answering | BioASQ | 14 | May 12, 2026 | ||
| Capacitated Vehicle Routing Problem | CVRP N=1000 (test) | 14 | May 12, 2026 | ||
| Capacitated Vehicle Routing Problem | CVRP N=500 (test) | 14 | May 12, 2026 | ||
| Capacitated Vehicle Routing Problem | CVRP N=200 (test) | 14 | May 12, 2026 | ||
| AI-generated Image Detection | ImageNet | 14 | May 12, 2026 | ||
| Causal Tool Verification | Causal-ToolBench Moderate Confounding n = 1,050 | 14 | May 12, 2026 |