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 |
|---|---|---|---|---|---|
| Motion Segmentation | Hopkins 155 (all sequences) | 57 | Jun 5, 2026 | ||
| Motion Segmentation | Hopkins 155 3-motion sequences | 57 | Jun 5, 2026 | ||
| Classification | AudioSet (test) | 57 | Feb 26, 2026 | ||
| Financial Question Answering | FinQA (test) | 57 | Mar 10, 2026 | ||
| Mathematical Reasoning | OlympiadBench | 57 | Mar 4, 2026 | ||
| Few-shot 3D Object Classification (5-way) |
|---|
| ModelNet40 (test) |
| 57 |
| Feb 26, 2026 |
| Robotic Manipulation | LIBERO 1.0 (test) | 57 | May 26, 2026 |
|---|
| Arithmetic Reasoning | AQUA | 57 | May 27, 2026 |
|---|
| 3D Point Cloud Generation | ShapeNet Car (test) | 57 | Feb 26, 2026 |
|---|
| Image Registration | OASIS (test) | 57 | Mar 19, 2026 |
|---|
| Question Answering | MMLU-Redux | 57 | May 26, 2026 |
|---|
| Semantic Segmentation | PASCAL 1-shot 5i | 57 | Feb 26, 2026 |
|---|
| Image Classification | CIFAR-10 | 56 | Jul 9, 2026 |
|---|
| Image Retrieval | IMAGENET@1000 | 56 | Jul 7, 2026 |
|---|
| Image Retrieval | CIFAR-10 @54000 | 56 | Jul 7, 2026 |
|---|
| Corruption Detection (Feature-noise) | ImageNet100 (test) | 56 | Jul 2, 2026 |
|---|
| Model Merging | Llama-3.2-1B-Instruct Task Set | 56 | Jul 2, 2026 |
|---|
| LoRA Merging | Qwen-1.5B-Instruct 2.5 (test) | 56 | Jul 2, 2026 |
|---|
| Visual SLAM | EuRoC (All Sequences) | 56 | Jul 2, 2026 |
|---|
| Backdoor Detection | SST2 | 56 | Jul 1, 2026 |
|---|
| LLM Judge Reliability Assessment | Therapeutic Dialogue Evaluation Set | 56 | Jul 1, 2026 |
|---|
| Deepfake Detection | GANs Out-of-domain (test) | 56 | Jun 30, 2026 |
|---|
| Multi-objective Optimization | DTLZ | 56 | Jun 30, 2026 |
|---|
| Evidence Retrieval | Synthetic Benchmark 1.0 (test) | 56 | Jun 30, 2026 |
|---|
| Multimodal Understanding | MLLM Evaluation Suite GQA, ScienceQA-IMG, TextVQA, POPE, MME, MMBench, MM-Vet, MMStar, AI2D, HallusionBench (test val) | 56 | Jun 26, 2026 |
|---|