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 |
|---|---|---|---|---|---|
| Image-to-Text Retrieval | MS-COCO 1K | 29 | Jul 3, 2026 | ||
| Node Classification | COCO-SP | 29 | Jun 4, 2026 | ||
| Sentence-to-Image Retrieval | MS-COCO 1K (5-folds) (test) | 29 | Feb 26, 2026 | ||
| Image-to-Sentence Retrieval | MS-COCO 1K (5-folds) (test) | 29 | Feb 26, 2026 | ||
| Referring Segmentation | refCOCOg U (test) | 29 | Jul 8, 2026 |
| Generalized Referring Expression Segmentation | gRefCOCO v1 (test B) | 29 | Feb 26, 2026 |
|---|
| Semantic Segmentation | COCO 1849 labels (1/64) | 29 | Feb 26, 2026 |
|---|
| Object Detection | MS-COCO (minival) | 29 | Mar 17, 2026 |
|---|
| Image Captioning | MSCOCO (test) | 29 | Feb 26, 2026 |
|---|
| Text Retrieval | MSCOCO (5K) | 29 | Feb 26, 2026 |
|---|
| Object Counting | COCO Count (test) | 29 | Feb 26, 2026 |
|---|
| AI-generated Video Detection | CoCoVideo (test) | 28 | Jun 2, 2026 |
|---|
| Object Hallucination Assessment in Image Captioning | MSCOCO 2014 (val) | 28 | May 26, 2026 |
|---|
| Overclustering | COCO Things | 28 | Apr 28, 2026 |
|---|
| Open-Vocabulary Object Detection | OV-COCO (test) | 28 | Apr 6, 2026 |
|---|
| Hallucination Evaluation | CHAIR MSCOCO 2014 | 28 | Apr 1, 2026 |
|---|
| Referring Object Detection | RefCOCO (testB) | 28 | Mar 26, 2026 |
|---|
| Referring Object Detection | RefCOCO (testA) | 28 | Mar 26, 2026 |
|---|
| Referring Object Detection | RefCOCO (val) | 28 | Mar 26, 2026 |
|---|
| Text-to-Image Generation | MS-COCO Stable Diffusion v1.5 (test) | 28 | Mar 24, 2026 |
|---|
| Image-Text Retrieval | MS-COCO 5K (test) | 28 | Apr 21, 2026 |
|---|
| Multi-label class-incremental learning | MS-COCO 2014 (B0-C10) | 28 | Mar 9, 2026 |
|---|
| Hallucination Evaluation | COCO | 28 | Mar 6, 2026 |
|---|
| Grounding | RefCOCO | 28 | Mar 4, 2026 |
|---|
| Style Erasure | MS-COCO | 28 | Mar 4, 2026 |
|---|