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 |
|---|---|---|---|---|---|
| Offline Reinforcement Learning | D4RL hopper-random | 86 | May 11, 2026 | ||
| Natural Language Inference | SciTail (test) | 86 | May 14, 2026 | ||
| Conditional Image Generation | ImageNet-1K 256x256 (val) | 86 | Feb 26, 2026 | ||
| Multi-image Reasoning | MIRB |
| 86 |
| Jun 23, 2026 |
| Text-to-Image Generation | T2I-CompBench (test) | 86 | Apr 10, 2026 |
|---|
| Image Classification | Caltech-256 (test) | 86 | Jun 12, 2026 |
|---|
| Table Question Answering | WikiTableQuestions (test) | 86 | Feb 26, 2026 |
|---|
| Named Entity Recognition | Conll 2003 | 86 | Feb 27, 2026 |
|---|
| Policy Composition | DexCompose 16 Composite Tasks 1.0 | 85 | Jun 29, 2026 |
|---|
| Zero-shot Evaluation | 9 diverse tasks zero-shot | 85 | May 27, 2026 |
|---|
| Audio Watermark Attribution | MusicCaps (test) | 85 | May 25, 2026 |
|---|
| Audio Watermark Detection | MusicCaps (test) | 85 | May 25, 2026 |
|---|
| Audio Watermark Detection | MusicCaps balanced (val) | 85 | May 25, 2026 |
|---|
| Zero-shot Evaluation | Zero-shot Evaluation Suite (ARC, Hellaswag, LAMBADA, PIQA, Winogrande) | 85 | May 20, 2026 |
|---|
| Image Classification | MNIST VAE m=32 latent space (test) | 85 | May 19, 2026 |
|---|
| Counterfactual Generation | SIB200 | 85 | May 13, 2026 |
|---|
| Common Sense Reasoning | HellaSwag | 85 | Jul 9, 2026 |
|---|
| Noise Multiplier Calculation | DP-SGD Hyperparameter Configurations (train) | 85 | May 11, 2026 |
|---|
| OOD Detection | ImageNet-54 (test) | 85 | Apr 24, 2026 |
|---|
| Visual Spatial Intelligence | VSI-Bench | 85 | Jul 3, 2026 |
|---|
| Image Classification | ImageIDC (test) | 85 | Jun 23, 2026 |
|---|
| Anomaly Generation | MVTec-AD | 85 | Mar 20, 2026 |
|---|
| Multi-hop Question Answering | 2WikiMHQA | 85 | Jun 5, 2026 |
|---|
| Physical Commonsense Reasoning | PIQA | 85 | Jun 12, 2026 |
|---|
| Robotic Manipulation | LIBERO (test) | 85 | Jul 7, 2026 |
|---|