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 |
|---|---|---|---|---|---|
| Dimensionality Reduction | Watch (test) | 40 | Jun 29, 2026 | ||
| Operations Research | BWOR | 40 | Jun 29, 2026 | ||
| Operations Research | IndustryOR | 40 | Jun 29, 2026 | ||
| Lattice Reduction | Subset Sum instances 20-bit weights | 40 | Jun 26, 2026 | ||
| Surveillance Reporting | SURVEILBENCH (All (303)) | 40 | Jun 25, 2026 | ||
| Structural Certification Filtering |
| Atari-based environment (|S|=100, |A|=5, T=20000) (test) |
| 40 |
| Jun 24, 2026 |
| Selective capability unlearning | SLURP Retain Set 1.0 (DR) | 40 | Jun 24, 2026 |
|---|
| Image Classification | SUN397 | 40 | Jun 30, 2026 |
|---|
| Training Efficiency Profiling | Qwen3 profiling S=2048 (train) | 40 | Jun 23, 2026 |
|---|
| Super-Resolution | Urban100 | 40 | Jun 23, 2026 |
|---|
| Super-Resolution | B100 | 40 | Jun 23, 2026 |
|---|
| Super-Resolution | Set14 | 40 | Jun 23, 2026 |
|---|
| Super-Resolution | Set5 | 40 | Jun 23, 2026 |
|---|
| Reproducibility Evaluation | MCR-Bench | 40 | Jun 23, 2026 |
|---|
| Physical Adversarial Attack | UAV Reconnaissance Video Dataset | 40 | Jun 19, 2026 |
|---|
| Multi-task model merging | 20 vision tasks | 40 | Jun 18, 2026 |
|---|
| Decoder Reconstruction | GEMMA-3-4B-IT | 40 | Jun 18, 2026 |
|---|
| Video Hallucination Evaluation | ModeVent (Overall) | 40 | Jun 17, 2026 |
|---|
| Video Hallucination Evaluation | ModeVent (Inliers) | 40 | Jun 17, 2026 |
|---|
| Question Answering | PubMedQA | 40 | Jun 25, 2026 |
|---|
| Visual Navigation | Real-world navigation Multiple-obstacle scenario | 40 | Jun 16, 2026 |
|---|
| Visual Navigation | Single-obstacle scenario Real-world navigation (10 sequences) | 40 | Jun 16, 2026 |
|---|
| Sleep stage classification | Sleep-EDF 5-Modal (test) | 40 | Jun 16, 2026 |
|---|
| Object Removal | Imagenette Other Classes v1 (test) | 40 | Jun 15, 2026 |
|---|
| Object Removal | Imagenette Erased Class v1 (test) | 40 | Jun 15, 2026 |
|---|