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 |
|---|---|---|---|---|---|
| Few-Shot Image Generation | AFHQ-Dog 10-shot | 34 | Feb 26, 2026 | ||
| Few-Shot Image Generation | AFHQ-Cat 10-shot | 34 | Feb 26, 2026 | ||
| Instance Segmentation | LVIS v1 (val) | 34 | Feb 26, 2026 | ||
| Generalized Few-Shot Semantic Segmentation | PASCAL 5^i (test) | 34 | Feb 26, 2026 | ||
| Anomaly Detection | ShanghaiTech (test) | 34 | Feb 26, 2026 | ||
| Depth Super-Resolution |
|---|
| Middlebury 2005-2014 (test) |
| 34 |
| Feb 26, 2026 |
| Semantic Segmentation | COCO-Object with background class | 34 | Apr 23, 2026 |
|---|
| Novel View Synthesis | ScanNet (test) | 34 | Mar 13, 2026 |
|---|
| Zero-Shot Image Classification | CUB | 34 | Feb 26, 2026 |
|---|
| Glaucoma Classification | retinal Glaucoma dataset (test) | 34 | Jun 16, 2026 |
|---|
| Classification | Phishing (test) | 34 | Jul 1, 2026 |
|---|
| Commonsense Reasoning | Commonsense Reasoning (BoolQ, PIQA, HellaSwag, Winogrande) zero-shot | 34 | Feb 26, 2026 |
|---|
| Query-based meeting summarization | QMSum (test) | 34 | Mar 5, 2026 |
|---|
| Node Classification | Actor (60%/20%/20% random splits) | 34 | Feb 26, 2026 |
|---|
| Unsupervised Single Object Discovery | VOC 2012 (test) | 34 | Feb 26, 2026 |
|---|
| Unsupervised Single Object Discovery | VOC 2007 (test) | 34 | Feb 26, 2026 |
|---|
| Person Re-identification | LTCC Standard setting | 34 | Mar 4, 2026 |
|---|
| Grammatical Error Correction | MuCGEC (test) | 34 | Feb 26, 2026 |
|---|
| Causal Inference | Demand (test) | 34 | Feb 26, 2026 |
|---|
| Super-resolution | Coltea-Lung-CT-100W 1.0 (test) | 34 | Feb 26, 2026 |
|---|
| Reading Comprehension | BoolQ (val) | 34 | Feb 26, 2026 |
|---|
| Commonsense Reasoning | StoryCloze | 34 | Feb 26, 2026 |
|---|
| Language Modeling | WikiText-103 zero-shot (test) | 34 | Feb 26, 2026 |
|---|
| Thoracic Disease Classification | MIMIC-CXR (test) | 34 | May 18, 2026 |
|---|
| 3D dense captioning | Nr3D | 34 | Jul 8, 2026 |
|---|