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 |
|---|---|---|---|---|---|
| Referring Expression Segmentation | RefCOCO+ UMD (testB) | 34 | May 28, 2026 | ||
| Referring Expression Segmentation | RefCOCO+ UMD (testA) | 34 | May 28, 2026 | ||
| Referring Expression Segmentation | RefCOCO+ UMD (val) | 34 | May 28, 2026 | ||
| Referring Expression Segmentation | RefCOCO UMD (testB) | 34 | May 28, 2026 | ||
| Mathematical Reasoning | DeepScaleR |
| 34 |
| Jun 30, 2026 |
| Image Classification | MNIST (test) | 34 | Jul 7, 2026 |
|---|
| Overall Performance | Aggregate 9 benchmarks | 34 | Jul 3, 2026 |
|---|
| Classification | OCTMNIST | 34 | May 27, 2026 |
|---|
| Classification | LC25000 | 34 | May 27, 2026 |
|---|
| Classification | CTKIDNEY | 34 | May 27, 2026 |
|---|
| Proactive Action Prediction | ABCD+ | 34 | May 26, 2026 |
|---|
| Multi-turn Dialogue | MT-Bench | 34 | May 26, 2026 |
|---|
| Joint Entity and Relation Extraction | WebNLG | 34 | May 25, 2026 |
|---|
| Dialogue | IFEval | 34 | May 22, 2026 |
|---|
| Dialogue | AlpacaEval 2 | 34 | May 22, 2026 |
|---|
| Myocardial Infarction Localization | Myocardial Infarction Localization (internal) | 34 | May 22, 2026 |
|---|
| Text-to-Image Generation | InitNO | 34 | May 20, 2026 |
|---|
| Multi-task Unlearning Interference | NYU v2 | 34 | May 20, 2026 |
|---|
| Continual Learning | KQA → Math → Code → MQA Task-level stream | 34 | May 19, 2026 |
|---|
| Continual Learning | Multi-hop QA Domain-level stream | 34 | May 19, 2026 |
|---|
| Continual Learning | Code Domain-level stream | 34 | May 19, 2026 |
|---|
| Continual Learning | Math Domain-level stream | 34 | May 19, 2026 |
|---|
| Continual Learning | Knowledge QA Domain-level stream | 34 | May 19, 2026 |
|---|
| Visual Question Answering | MMStar (test) | 34 | Jun 9, 2026 |
|---|
| Text-to-Video Generation | EvalCrafter | 34 | May 19, 2026 |
|---|