Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 1487 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Image Classification | SUN397 | 21 | Apr 23, 2026 | ||
| Semantic Segmentation | PASCAL VOC PAS-20 foreground categories (val) | 21 | Feb 26, 2026 | ||
| Image Classification | Pattern | 21 | Feb 26, 2026 | ||
| Image Classification | 12 Downstream Datasets Fine-grained, Natural, Specialized (test) | 21 | Feb 26, 2026 | ||
| Long-video understanding | VNBench | 21 | Feb 26, 2026 | ||
| Visual Question Answering | CameraBench | 21 | Mar 20, 2026 | ||
| Text-based Visual Question Answering | TextVQA | 21 | May 27, 2026 | ||
| Adversarial Prompt Attack | Parachute concept | 21 | Feb 26, 2026 | ||
| Adversarial Prompt Attack | Church concept | 21 | Feb 26, 2026 | ||
| Adversarial Prompt Attack | Van Gogh concept | 21 | Feb 26, 2026 | ||
| Adversarial Prompt Attack | Nudity concept | 21 | Feb 26, 2026 | ||
| Multimodal Large Language Model Inference Efficiency | LLaVA 13B 1.5 (test) | 21 | Feb 26, 2026 | ||
| Indoor Occupancy Prediction | Occ-ScanNet (val) | 21 | May 19, 2026 | ||
| Personalized Image Generation | Personalized Image Generation dataset | 21 | Feb 26, 2026 | ||
| Video Grounding | ActivityNet Captions 1.3 (test val) | 21 | Feb 26, 2026 | ||
| Image Classification | Caltech256 | 21 | Apr 14, 2026 | ||
| Object Detection | Pascal-5i 2010 (Novel Split 2) | 21 | Feb 27, 2026 | ||
| Cardiac Image Segmentation | M&MS (test) | 21 | Mar 27, 2026 | ||
| Hyperspectral Anomaly Detection | Pavia | 21 | Feb 27, 2026 | ||
| Out-of-Distribution Detection | CIFAR-100 In-Dist FarOOD Aggregate | 21 | Mar 25, 2026 | ||
| Object Presence Recognition | OPOPE | 21 | Feb 26, 2026 | ||
| Class-incremental learning | ImageNet-R 20-split | 21 | Feb 26, 2026 | ||
| Skeleton Action Recognition | PKU Part II (test) | 21 | Feb 26, 2026 | ||
| Text-to-Speech | Seed-TTS-Eval zh (test) | 21 | May 26, 2026 | ||
| Temporal Grounding | ActivityNet-MR | 21 | May 27, 2026 |