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 |
|---|---|---|---|---|---|
| 3D Gaussian Splatting Compression | DeepBlending | 36 | May 21, 2026 | ||
| 3D Gaussian Splatting Compression | Mip-NeRF360 | 36 | May 21, 2026 | ||
| Multi-modal Reasoning | MMMU-Pro | 36 | Apr 23, 2026 | ||
| Math Problem Solving | 200 IMO-level math problems IMO-AnswerBench, IMO-ProofBench, ArXivMath (test) | 36 | Mar 31, 2026 | ||
| Image Captioning | COCO |
| 36 |
| Mar 31, 2026 |
| Image Classification | ImageNet-R | 36 | Jun 2, 2026 |
|---|
| Time Series Anomaly Detection | MSL | 36 | May 12, 2026 |
|---|
| Mathematical Reasoning | MMATH | 36 | Jun 2, 2026 |
|---|
| Code Generation | MBPP | 36 | Apr 6, 2026 |
|---|
| Spatio-temporal data upscaling | Experiment Set 1 | 36 | Mar 26, 2026 |
|---|
| Dummy-Target Retrieval | FMA | 36 | Mar 26, 2026 |
|---|
| Position error correction | GP01 receiver 3×CW jamming | 36 | Mar 25, 2026 |
|---|
| Position error correction | GP01 receiver CW jamming | 36 | Mar 25, 2026 |
|---|
| Model Poisoning Defense | FHWA | 36 | Mar 25, 2026 |
|---|
| Model Poisoning Defense | PDCCH | 36 | Mar 25, 2026 |
|---|
| Embodied Video Generation | PAI-Bench robot domain | 36 | Jun 29, 2026 |
|---|
| Jailbreak Attack | AdvBench | 36 | Mar 25, 2026 |
|---|
| Semantic Textual Similarity | SICK-R (test) | 36 | Jun 16, 2026 |
|---|
| Near-OOD Detection | ImageNet-200 | 36 | Apr 15, 2026 |
|---|
| Full Waveform Inversion | Overthrust model | 36 | Mar 25, 2026 |
|---|
| Full Waveform Inversion | BP model 2004 | 36 | Mar 25, 2026 |
|---|
| Jailbreak Defense | LLaVA v1.5 | 36 | Mar 24, 2026 |
|---|
| Toxicity Defense | LLaVA v1.5 | 36 | Mar 24, 2026 |
|---|
| Jailbreak Defense | Qwen2-VL | 36 | Mar 24, 2026 |
|---|
| Toxicity Defense | Qwen2-VL | 36 | Mar 24, 2026 |
|---|