Loading the SOTA2 catalog…
SOTA2 Research · benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
| Task | Dataset | Domain | Trend | Results | Last Update |
|---|---|---|---|---|---|
| Multi-label image recognition | MS-COCO 2014 (val) | 51 | Feb 26, 2026 | ||
| Visual Question Answering | COCO-QA (test) | 51 | Mar 27, 2026 | ||
| Object Detection | MS COCO small objects (val) | 50 | Jun 2, 2026 | ||
| Referring Expression Segmentation | refCOCO UMD (val) | 50 | Feb 26, 2026 | ||
| Object Hallucination |
| MSCOCO 500 images 2014 (val) |
| 50 |
| Feb 26, 2026 |
| Object Detection | MS COCO 2017 (minival) | 50 | Feb 26, 2026 |
|---|
| Object Detection | COCO | 49 | May 19, 2026 |
|---|
| Referring Expression Comprehension | RefCOCO v1 (val) | 49 | Feb 26, 2026 |
|---|
| Text-to-Image Retrieval | COCO-CN | 49 | Feb 26, 2026 |
|---|
| Semantic Image Synthesis | COCO-Stuff | 49 | Mar 24, 2026 |
|---|
| Object Detection | COCO mini 2017 (val) | 49 | Feb 26, 2026 |
|---|
| Instance Segmentation | COCO | 48 | May 13, 2026 |
|---|
| Object Detection | COCO | 48 | May 13, 2026 |
|---|
| Text-to-Image Generation | MS-COCO zero-shot 512 x 512 Stable Diffusion v3.5 | 48 | Feb 26, 2026 |
|---|
| Image Generation | MS-COCO 512x512 with Stable Diffusion (val) | 48 | Feb 26, 2026 |
|---|
| Video Jailbreaking | COCO 2017 (test) | 48 | Feb 26, 2026 |
|---|
| Image Matching | COCO (val) | 48 | Feb 26, 2026 |
|---|
| Semantic Segmentation | COCO 20^i (test) | 48 | Feb 26, 2026 |
|---|
| Image-to-Text Retrieval | COCO-CN | 48 | Feb 26, 2026 |
|---|
| Style Transfer | MS-COCO and WikiArt | 48 | Feb 26, 2026 |
|---|
| Text-to-Image Retrieval | MSCOCO | 47 | Jun 2, 2026 |
|---|
| Semantic Segmentation | COCO-Stuff-10K (test) | 47 | Feb 26, 2026 |
|---|
| Semantic Segmentation | COCOStuff 164k (val) | 47 | Apr 8, 2026 |
|---|
| Object Detection | COCO-FSOD 30-shot COCO-20 | 47 | Feb 26, 2026 |
|---|
| Object Hallucination Detection | MSCOCO | 46 | May 13, 2026 |
|---|