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 |
|---|---|---|---|---|---|
| Mathematical Reasoning | GPQA Diamond (test) | 54 | Jun 1, 2026 | ||
| Mathematical Reasoning | AMC 2025 (test) | 54 | Jun 1, 2026 | ||
| Embodied Task Completion | ALFWorld | 54 | Jun 26, 2026 | ||
| Mathematical Reasoning | GSM8K (test) | 54 | Jul 1, 2026 | ||
| Text Generation | OpenWebText | 54 | Jun 1, 2026 |
| Offline Reinforcement Learning | Ant-Kinematic | 54 | May 22, 2026 |
|---|
| Offline Reinforcement Learning | Halfcheetah-Morph | 54 | May 22, 2026 |
|---|
| Regression | Synthetic high-dimensional data (test) | 54 | May 20, 2026 |
|---|
| Object Detection | BDD-100k | 54 | May 19, 2026 |
|---|
| Object Detection | MSCOCO | 54 | May 19, 2026 |
|---|
| Topic Detection | Text Datasets active topics | 54 | May 19, 2026 |
|---|
| Hypergraph Parameter Estimation | Synthetic Hypergraph K=8, rho in [0,1] | 54 | May 19, 2026 |
|---|
| Synthetic hypergraph generation | Synthetic Hypergraph K=2, rho_i in [1, 2] | 54 | May 19, 2026 |
|---|
| Prediction-grounded correlation with accuracy difference | SST-2 | 54 | May 18, 2026 |
|---|
| Long-video Understanding | VideoNIAH | 54 | May 15, 2026 |
|---|
| Long-video Understanding | LSDBench | 54 | May 15, 2026 |
|---|
| Image Classification | SVHN | 54 | Jun 30, 2026 |
|---|
| Joint Replenishment Problem | JRP (test) | 54 | May 15, 2026 |
|---|
| Natural Language Understanding | GLUE (val) | 54 | Jul 3, 2026 |
|---|
| Ultrasound Image Denoising | Simulation Ultrasound Data | 54 | May 14, 2026 |
|---|
| Document Visual Question Answering | DocVQA | 54 | Jul 7, 2026 |
|---|
| Logical Reasoning | SLR-BENCH Extended Leaderboard | 54 | May 12, 2026 |
|---|
| Multi-step tool use execution | M3ToolEval | 54 | May 12, 2026 |
|---|
| Depth Super-Resolution | TOFDSR (test) | 54 | May 12, 2026 |
|---|
| Depth Super-Resolution | RGB-D-D (test) | 54 | May 12, 2026 |
|---|