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 |
|---|---|---|---|---|---|
| Action Recognition | UCF-101 Few-shot | 30 | Feb 26, 2026 | ||
| Temporal Action Detection | HACS segment (test) | 30 | Feb 26, 2026 | ||
| Action Recognition | HMDB51 | 30 | Feb 26, 2026 | ||
| Fact Verification | RAWFC | 30 | Feb 26, 2026 | ||
| Knowledge Retrieval / Relation Prediction | Person-Instrument | 30 | Feb 27, 2026 | ||
| Mathematical Reasoning |
| GSM8K Numerical +4 (test) |
| 30 |
| Feb 26, 2026 |
| Visual Perception | MMStar | 30 | May 20, 2026 |
|---|
| Perception | MMStar latest (test) | 30 | Feb 26, 2026 |
|---|
| Procedure Planning | NIV T=3 (test) | 30 | Mar 11, 2026 |
|---|
| Model Extraction | Randomly generated models Layer 2 | 30 | Feb 26, 2026 |
|---|
| Texas Hold'em | Texas Hold'em | 30 | Feb 26, 2026 |
|---|
| Category-Agnostic Pose Estimation | MP-100 (Average) | 30 | May 14, 2026 |
|---|
| Category-Agnostic Pose Estimation | MP-100 (Split 5) | 30 | May 14, 2026 |
|---|
| Category-Agnostic Pose Estimation | MP-100 (Split 4) | 30 | May 14, 2026 |
|---|
| Category-Agnostic Pose Estimation | MP-100 (Split 3) | 30 | May 14, 2026 |
|---|
| Category-Agnostic Pose Estimation | MP-100 (Split 2) | 30 | May 14, 2026 |
|---|
| Text-based Person Retrieval | ICFG-PEDES (test) | 30 | Mar 31, 2026 |
|---|
| Diabetic Retinopathy Classification | DeepDR (test) | 30 | Feb 26, 2026 |
|---|
| Model Editing | CounterFact | 30 | Feb 26, 2026 |
|---|
| Exemplar-Free Class-Incremental Learning | CIFAR-100 (test) | 30 | Feb 26, 2026 |
|---|
| Exemplar-Free Class Incremental Learning | ImageNet Subset (test) | 30 | Feb 26, 2026 |
|---|
| Class Incremental Object Detection | COCO 2017 (val) | 30 | Feb 26, 2026 |
|---|
| Bigram Language Modeling | Synthetic WebText initialization (val) | 30 | Feb 26, 2026 |
|---|
| Bigram Language Modeling | Synthetic Random 50% initialization (val) | 30 | Feb 26, 2026 |
|---|
| Sentiment Transformation | IMDB PosToNeg 1.0 (test) | 30 | Feb 26, 2026 |
|---|