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 |
|---|---|---|---|---|---|
| Personalized Question Answering | PFQABench 1.0 (test) | 48 | Feb 26, 2026 | ||
| Membership Inference Attack | MIMIR Average | 48 | Apr 1, 2026 | ||
| Mathematical Reasoning | MATH OOD | 48 | Jun 4, 2026 | ||
| Scientific Reasoning | GPQA Diamond | 48 | Apr 3, 2026 | ||
| Multi-hop Question Answering | HotpotQA | 48 |
| Feb 26, 2026 |
| Mathematical Reasoning | MATH | 48 | Feb 26, 2026 |
|---|
| Mathematical Reasoning | GSM8K (test) | 48 | Feb 26, 2026 |
|---|
| Hallucination Detection | Average Cross-domain | 48 | Feb 26, 2026 |
|---|
| Hallucination Detection | Belebele | 48 | Feb 26, 2026 |
|---|
| GUI navigation | AndroidWorld latest (test) | 48 | Jul 1, 2026 |
|---|
| Mathematical Reasoning | AIME25 | 48 | May 13, 2026 |
|---|
| Question Answering | TriviaQA | 48 | Feb 26, 2026 |
|---|
| Reasoning length evaluation | 20 attack prompts | 48 | Feb 26, 2026 |
|---|
| Output Length Prediction | ForeLen LongSeq | 48 | Feb 26, 2026 |
|---|
| Prompt-based Value Alignment | Schwartz Value Prompts (test) | 48 | Feb 26, 2026 |
|---|
| Error Detection | EuroSAT | 48 | Apr 28, 2026 |
|---|
| Question Answering | AVG. | 48 | May 19, 2026 |
|---|
| Moving Object Segmentation | DAVIS Moving 2016 | 48 | Jun 30, 2026 |
|---|
| Click-Through Rate Prediction | KKBox | 48 | Feb 26, 2026 |
|---|
| 3D Question Answering | ScanQA | 48 | Jul 7, 2026 |
|---|
| LiDAR Localization | NCLT Average | 48 | Jun 29, 2026 |
|---|
| Video Part Segmentation | VIP | 48 | Jun 23, 2026 |
|---|
| Scene Graph Anticipation | Action Genome AGS F=0.7 | 48 | Feb 26, 2026 |
|---|
| Object Detection | PASCAL-VOC 07+12 | 48 | Feb 26, 2026 |
|---|
| Talking Head Generation | HDTF | 48 | May 19, 2026 |
|---|