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 |
|---|---|---|---|---|---|
| Text-based embodied AI | ALFWorld | 30 | May 15, 2026 | ||
| Medical Reasoning | DiSCQ | 30 | May 15, 2026 | ||
| Medical Reasoning | Time-MMD | 30 | May 15, 2026 | ||
| Medical Reasoning | MedQA | 30 | May 15, 2026 | ||
| Jailbreak Defense | MaliciousInstruct | 30 | May 15, 2026 | ||
| Jailbreak Defense |
| JailbreakBench |
| 30 |
| May 15, 2026 |
| Summarization | SAMSum | 30 | May 15, 2026 |
|---|
| Closed-domain QA | BoolQ | 30 | May 15, 2026 |
|---|
| Random Band Masking Reconstruction | Pavia University (test) | 30 | May 15, 2026 |
|---|
| Question Answering | SciQA (test) | 30 | May 15, 2026 |
|---|
| Code Generation | CodeContests | 30 | May 29, 2026 |
|---|
| Question Answering | HotpotQA | 30 | May 15, 2026 |
|---|
| Mathematical Reasoning | MATH | 30 | May 15, 2026 |
|---|
| Mathematical Reasoning | GSM8K | 30 | May 15, 2026 |
|---|
| General Competence | G2U Overall | 30 | May 15, 2026 |
|---|
| Utility | G2U Utility | 30 | May 15, 2026 |
|---|
| Curiosity | G2U Curiosity | 30 | May 15, 2026 |
|---|
| Survival | G2U Survival | 30 | May 15, 2026 |
|---|
| Watermark Removal | Watermarked Text 1500 tokens | 30 | May 15, 2026 |
|---|
| Watermark Removal | Watermarked Text 500 tokens | 30 | May 15, 2026 |
|---|
| Prompt-to-prompt semantic composition | GenAI-Bench | 30 | May 15, 2026 |
|---|
| Deformable registration | HCP T2w–T1w (test) | 30 | May 14, 2026 |
|---|
| Deformable registration | Abdomen MR–CT (L2OCV folds) | 30 | May 14, 2026 |
|---|
| Polyp Segmentation | CVC-ClinicDB, CVC-ColonDB, and Kvasir-SEG (Macro-averaged) | 30 | May 19, 2026 |
|---|
| Hyperspectral Image Restoration | IH-10 1.0 (test) | 30 | May 14, 2026 |
|---|