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 |
|---|---|---|---|---|---|
| Molecular property prediction | BBBP 50-shot (test) | 40 | May 18, 2026 | ||
| Composed Image Retrieval | CIRR Subset (test) | 40 | Jun 18, 2026 | ||
| 3D Semantic Segmentation | S3DIS (S0, S1) | 40 | Feb 26, 2026 | ||
| Semantic Correspondence | SPair-71k | 40 | Jul 7, 2026 | ||
| Super-Resolution | BSD100 x2 (test) | 40 | May 26, 2026 | ||
| Multi-view Depth Estimation |
| DDAD (test) |
| 40 |
| Feb 26, 2026 |
| Motion completion | HumanML3D (test) | 40 | Feb 26, 2026 |
|---|
| Trajectory Prediction | nuScenes 1.0 (val) | 40 | May 13, 2026 |
|---|
| CT Report Generation | CTRG-Chest-548K (test) | 40 | Jun 2, 2026 |
|---|
| Multi-modal Understanding | MMBench (dev) | 40 | Feb 27, 2026 |
|---|
| Semantic Change Detection | Landsat-SCD | 40 | Jun 9, 2026 |
|---|
| Cancer Subtyping | BRACS-7 | 40 | May 20, 2026 |
|---|
| Language Modeling | Arxiv Proof-pile | 40 | Mar 4, 2026 |
|---|
| Image Classification | ImageNet-1k 1.0 (val) | 40 | Feb 26, 2026 |
|---|
| Univariate Time Series Forecasting | weather | 40 | Jun 8, 2026 |
|---|
| Group Activity Recognition | Volleyball dataset | 40 | Feb 27, 2026 |
|---|
| Synthetic Image Detection | GANs dataset | 40 | May 26, 2026 |
|---|
| Collaborative Filtering | Gowalla | 40 | Feb 26, 2026 |
|---|
| Motion Magnification | Real-world Videos (test) | 40 | Feb 26, 2026 |
|---|
| Multi-modal Understanding | SEED-Bench (overall) | 40 | Feb 26, 2026 |
|---|
| Text Classification | MNLI | 40 | Jun 4, 2026 |
|---|
| Object Detection | FLIR Aligned (test) | 40 | May 14, 2026 |
|---|
| Open-ended | AlpacaEval | 40 | Mar 24, 2026 |
|---|
| No-Reference Video Quality Assessment | LSVQ (test) | 40 | Feb 26, 2026 |
|---|
| Interpretation Error | ImageNet (test) | 40 | Feb 26, 2026 |
|---|