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 |
|---|---|---|---|---|---|
| Channel prediction | Stochastic Velocity [10, 20) m/s | 48 | Mar 12, 2026 | ||
| Channel prediction | Stochastic Velocity [0, 10) m/s | 48 | Mar 12, 2026 | ||
| Channel prediction | Ray-Traced Velocity [20, 30) m/s | 48 | Mar 12, 2026 | ||
| Channel prediction | Ray-Traced Velocity [10, 20) m/s | 48 | Mar 12, 2026 | ||
| Channel prediction | Ray-Traced Velocity [0, 10) m/s |
| 48 |
| Mar 12, 2026 |
| General QA | TriviaQA | 48 | May 11, 2026 |
|---|
| Sliced Wasserstein Distance Estimation | Gaussian toy samples d=30 | 48 | Mar 11, 2026 |
|---|
| Long-term Forecasting | Weather | 48 | Jun 4, 2026 |
|---|
| Long-term Forecasting | Exchange Rate (ExRate) | 48 | Jun 4, 2026 |
|---|
| Heart Rate (HR) Forecasting | BIDMC PPG and Respiration (53 LOPO folds) | 48 | Mar 10, 2026 |
|---|
| Spatial Reasoning | SPBench SI | 48 | Jun 18, 2026 |
|---|
| Long-horizon robotic manipulation | CALVIN ABC-D | 48 | Jun 16, 2026 |
|---|
| Capacitated Vehicle Routing Problem | CVRP 20 | 48 | Jul 1, 2026 |
|---|
| Visual Reasoning | MMBench | 48 | Jun 9, 2026 |
|---|
| Emotion Recognition | EAV | 48 | Jul 7, 2026 |
|---|
| Multiresolution Rendering | Replica | 48 | Mar 10, 2026 |
|---|
| Machine Unlearning | CIFAR-100 | 48 | Mar 10, 2026 |
|---|
| Text-to-Speech | English emotional dataset | 48 | Mar 9, 2026 |
|---|
| Lidar Odometry | KITTI-odometry (sequences 00-10) | 48 | Apr 3, 2026 |
|---|
| Class-conditional generation | ImageNet 512x512 (test) | 48 | May 19, 2026 |
|---|
| Adversarial Attack | Flickr30K | 48 | Mar 6, 2026 |
|---|
| dynamic link prediction | Contact (transductive) | 48 | Jun 4, 2026 |
|---|
| Global Prediction | CIFAR-100 original uncorrupted (train) | 48 | Mar 4, 2026 |
|---|
| Personalization Performance | CIFAR-C-100 (9 held-out corruption types) | 48 | Mar 4, 2026 |
|---|
| Image Classification | CIFAR-C-100 10 training corruption types (test) | 48 | Mar 4, 2026 |
|---|