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 |
|---|---|---|---|---|---|
| Classification | bbc | 61 | May 26, 2026 | ||
| Out-of-Distribution Detection | FashionMNIST (ID) vs MNIST (OoD) | 61 | Feb 26, 2026 | ||
| Speaker Verification | VoxCeleb1-O Cleaned (Original) | 61 | Apr 29, 2026 | ||
| Music Source Separation | MUSDB18 HQ (test) | 61 | Apr 13, 2026 | ||
| Offline Reinforcement Learning |
| walker2d medium-replay |
| 61 |
| May 29, 2026 |
| Offline Reinforcement Learning | walker2d medium | 61 | May 29, 2026 |
|---|
| Common Sense Reasoning | WG | 61 | Jun 26, 2026 |
|---|
| Transductive Node Classification | Citeseer (transductive) | 61 | Feb 26, 2026 |
|---|
| Polyp Segmentation | EndoScene | 61 | Feb 26, 2026 |
|---|
| Commonsense Question Answering | CSQA | 61 | May 19, 2026 |
|---|
| Mathematical Reasoning | GaoKao (test) | 61 | Feb 26, 2026 |
|---|
| Referring Video Object Segmentation | A2D-Sentences | 61 | Apr 21, 2026 |
|---|
| Visual Question Answering | HRBench-4K | 61 | May 21, 2026 |
|---|
| Knowledge Distillation | Oxford Flowers 102 (test) | 60 | Jul 8, 2026 |
|---|
| Ranking | UltraFeedback clean (held-out) | 60 | Jul 3, 2026 |
|---|
| Program Repair | Real misspecified programs neutral briefs | 60 | Jul 1, 2026 |
|---|
| Watermark Forgery Detection | Stable-Diffusion-Prompts 1,000 samples (test) | 60 | Jun 30, 2026 |
|---|
| Exercise Assessment | UI-PRMD (test) | 60 | Jun 30, 2026 |
|---|
| Multi-Layer Spatial Relationship Accuracy | MD-3k (Overall) | 60 | Jun 30, 2026 |
|---|
| Object Hallucination Evaluation | A-OKVQA POPE SEEM-annotated | 60 | Jun 26, 2026 |
|---|
| Object Hallucination Evaluation | MSCOCO POPE SEEM-annotated | 60 | Jun 26, 2026 |
|---|
| Long-context language understanding | LongBench | 60 | Jun 24, 2026 |
|---|
| Anomaly Segmentation | MANTA Tiny (test) | 60 | Jun 24, 2026 |
|---|
| Object Detection | Pascal VOC Novel Split 1 | 60 | Jun 23, 2026 |
|---|
| End-to-End Selection | AlgoSimBench MCQ | 60 | Jun 23, 2026 |
|---|