Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 1938 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Image Classification | Flower (test) | 18 | Feb 26, 2026 | ||
| High Dynamic Range View Synthesis | HDR-Plenoxels Real Scenes LDR-NE (t2, t4) | 18 | Mar 31, 2026 | ||
| High Dynamic Range View Synthesis | HDR-Plenoxels Real Scenes LDR-OE (t1, t3, t5) | 18 | Mar 31, 2026 | ||
| Image Generation | LAION-COCO Vertical | 18 | Feb 26, 2026 | ||
| Image Generation | LAION-COCO Horizontal | 18 | Feb 26, 2026 | ||
| Image Classification | Places365 3-shot | 18 | Feb 26, 2026 | ||
| Cooperative Object Detection | OPV2V | 18 | Feb 26, 2026 | ||
| Cross-dataset generalization | 10 Target Datasets (Caltech, Pets, Cars, Flowers, Food101, Aircraft, SUN397, DTD, EuroSAT, UCF101) various (test) | 18 | Feb 26, 2026 | ||
| Image Classification | Biomedical Datasets Average (test) | 18 | Feb 26, 2026 | ||
| Adversarial Attack | DVS-Gesture binary (test) | 18 | Feb 26, 2026 | ||
| Classification | DTD | 18 | Feb 27, 2026 | ||
| Human Preference Alignment | MM-AlignBench 1.0 (test) | 18 | Feb 26, 2026 | ||
| Dialogue Generation | Proposed Multi-scenario Dataset 1.0 (test) | 18 | Feb 26, 2026 | ||
| Regression | AFAD B | 18 | Feb 26, 2026 | ||
| LLM Unlearning | MT-Bench | 18 | Feb 26, 2026 | ||
| LLM Unlearning | UNCD-Cyber (Retain) | 18 | Feb 26, 2026 | ||
| LLM Unlearning | UNCD-Cyber (Forget) | 18 | Feb 26, 2026 | ||
| Tabular Data Utility | Adult (test) | 18 | May 27, 2026 | ||
| Lesion Segmentation | Liver Tumor | 18 | Jun 29, 2026 | ||
| Citation Attribution | CNN-Daily Mail | 18 | Feb 26, 2026 | ||
| Visual Place Recognition | SF-XL (Night) | 18 | Feb 26, 2026 | ||
| Visual Place Recognition | SF-XL Occ. | 18 | Feb 26, 2026 | ||
| Commonsense Reasoning | Commonsense Reasoning LLaMA2-7B | 18 | Feb 26, 2026 | ||
| Mathematical Reasoning | AIME 2024 and 2025 (test) | 18 | Mar 24, 2026 | ||
| Single-view 3D Reconstruction | GSO (test) | 18 | May 6, 2026 |