Research
Benchmarks for “coco”
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
3,140 benchmarksPage 83 of 126
Search results
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Image Description | COCO 2017 (val) | 6 | Mar 16, 2026 | ||
| Visual Question Answering | COCO FuzzingRL probes (held-out) | 6 | Mar 10, 2026 | ||
| Object Recognition | COCO obj-80 | 6 | Mar 9, 2026 | ||
| Referring Expression Comprehension | RefCOCO+ | 6 | Mar 6, 2026 | ||
| Image Captioning | COCOCap | 6 | Mar 6, 2026 | ||
| Prompt Recovery | MS COCO | 6 | Mar 6, 2026 | ||
| Text-to-Image Generation | GenEval and MS-COCO 10K subset | 6 | Mar 4, 2026 | ||
| Text-to-Image Generation | COCO captions 2017 (test) | 6 | Mar 4, 2026 | ||
| Segmentation | COCONut (val) | 6 | Mar 4, 2026 | ||
| Few-concept Erasure (Single Instance) | MS-COCO Instance templates | 6 | Mar 4, 2026 | ||
| Multi-concept erasure | MS-COCO 100 celebrities erasure scenario | 6 | Mar 4, 2026 | ||
| Multi-concept erasure | MS-COCO 50 celebrities erasure scenario | 6 | Mar 4, 2026 | ||
| Multi-concept erasure | MS-COCO 10 celebrities erasure scenario | 6 | Mar 4, 2026 | ||
| Multi-label Classification | MS-COCO 2014 | 6 | Feb 27, 2026 | ||
| Object composition | COCO 2014 (val) | 6 | Feb 27, 2026 | ||
| Object Detection | MSCOCO Target 17 classes | 6 | Feb 27, 2026 | ||
| Bounding-box alignment | COCO | 6 | Feb 26, 2026 | ||
| Referring Localization | RefCOCO (val) | 6 | Feb 26, 2026 | ||
| Controlled Caption Generation | COCO Localized Narratives (test) | 6 | Feb 26, 2026 | ||
| Object Grounding | Refcoco+ (TestA) | 6 | Feb 26, 2026 | ||
| Feature Robustness | ImageNet, MSCOCO, and Caltech-256 (test) | 6 | Feb 26, 2026 | ||
| Image Classification | MSCOCO | 6 | Feb 26, 2026 | ||
| Object Proposal | COCO all categories 80 classes | 6 | Feb 26, 2026 | ||
| Concept Detection | COCO (test) | 6 | Feb 26, 2026 | ||
| Instance Segmentation | MS COCO 2017 (test) | 6 | Feb 26, 2026 |