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 |
|---|---|---|---|---|---|
| Science Question Answering | ScienceQA Image | 70 | May 14, 2026 | ||
| Visual Perception and Reasoning | BLINK | 70 | Jun 9, 2026 | ||
| OOD detection | ISIC Ink Artefacts (Similar) | 70 | Feb 27, 2026 | ||
| Uncertainty Quantification | Musique 500 randomly sampled queries (test) | 70 | Feb 26, 2026 | ||
| Uncertainty Quantification | HotpotQA 500 randomly sampled queries (test) |
| 70 |
| Feb 26, 2026 |
| Uncertainty Quantification | PopQA 500 randomly sampled queries (test) | 70 | Feb 26, 2026 |
|---|
| Image Embedding | MMEB v1 (test) | 70 | Apr 8, 2026 |
|---|
| Image Classification | MNIST | 70 | May 1, 2026 |
|---|
| Model Reduction | Shallow Water trajectory manifold of the KdV equation (test) | 70 | Feb 26, 2026 |
|---|
| Continual Concept Learning | 10 Sequential Concepts (test) | 70 | Feb 26, 2026 |
|---|
| Feature Selection | Simulated Data | 70 | Feb 26, 2026 |
|---|
| Idea Generation | LiveIdeaBench | 70 | Feb 26, 2026 |
|---|
| Jailbreak | JBB | 70 | Feb 26, 2026 |
|---|
| Jailbreak | Sorry | 70 | Feb 26, 2026 |
|---|
| Retrieval Attack Defense | FiQA | 70 | May 27, 2026 |
|---|
| Question Answering | NQ | 70 | Mar 4, 2026 |
|---|
| Emotion Recognition | SEED Session 1 → Session 2 | 70 | Feb 26, 2026 |
|---|
| Visible-Infrared Person Re-Identification | SYSU-MM01 All Search v1 | 70 | Feb 26, 2026 |
|---|
| Object Hallucination Evaluation | POPE GQA Popular | 70 | Jun 1, 2026 |
|---|
| NSFW Concept Generation | NSFW-200 Violence v2.1 (test) | 70 | Feb 26, 2026 |
|---|
| NSFW Concept Generation | NSFW-200 Sex v2.1 (test) | 70 | Feb 26, 2026 |
|---|
| Medical Question Answering | MedExpQA | 70 | Mar 5, 2026 |
|---|
| Reward Modeling | RewardBench Precise IF 2 | 70 | Feb 26, 2026 |
|---|
| Zero-shot Evaluation | 6 zero-shot downstream tasks | 70 | May 19, 2026 |
|---|
| Visual Logical Reasoning | LogicVista | 70 | Mar 31, 2026 |
|---|