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 |
|---|---|---|---|---|---|
| Expert Placement | Canonical multi-task workload | 32 | Jun 2, 2026 | ||
| Video Question Answering | VideoMME | 32 | Jul 7, 2026 | ||
| Speculative Decoding | Avg. | 32 | Jun 26, 2026 | ||
| Pointwise evaluation | BIGGEN | 32 | Jun 1, 2026 | ||
| Text Classification | GLUE | 32 | Jun 1, 2026 | ||
| ScanNet++ |
| 32 |
| Jul 7, 2026 |
| Depth Completion | ETH3D (test) | 32 | May 29, 2026 |
|---|
| EEG Super-Resolution | Localize-MI 256 ch. | 32 | May 29, 2026 |
|---|
| Reasoning | Math Vision | 32 | May 28, 2026 |
|---|
| Radio Environmental Mapping | RadioMapSeer | 32 | May 28, 2026 |
|---|
| Video Sentence Grounding | ActivityNet Captions | 32 | May 28, 2026 |
|---|
| Graph Anomaly Detection | T-Social | 32 | May 28, 2026 |
|---|
| Multimodal Dental Question Answering | MMOral-Uni | 32 | May 28, 2026 |
|---|
| Higher-frequency wave-field prediction | Maxwell (test) | 32 | May 27, 2026 |
|---|
| Higher-frequency wave-field prediction | Helmholtz (test) | 32 | May 27, 2026 |
|---|
| Higher-frequency wave-field prediction | SimpleWave (test) | 32 | May 27, 2026 |
|---|
| Uncertainty Estimation | Secret-word taboo dataset (random-split) | 32 | May 26, 2026 |
|---|
| Cross-lingual Retrieval | WebFAQ | 32 | May 26, 2026 |
|---|
| Logical Reasoning | K&K tasks | 32 | May 26, 2026 |
|---|
| Mathematical Reasoning | AMC 2023 | 32 | Jun 30, 2026 |
|---|
| Visual Prompt Injection Defense | ImgHijack Specific String (test) | 32 | May 26, 2026 |
|---|
| Offline Reinforcement Learning | MuJoCo Hopper Friction shift | 32 | May 26, 2026 |
|---|
| Offline Reinforcement Learning | MuJoCo Walker2d Friction shift | 32 | May 26, 2026 |
|---|
| Offline Reinforcement Learning | MuJoCo Walker2d Gravity shift | 32 | May 26, 2026 |
|---|
| Offline Reinforcement Learning | MuJoCo Ant Friction shift | 32 | May 26, 2026 |
|---|