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 |
|---|---|---|---|---|---|
| Selective Answering | integrated benchmark | 32 | May 20, 2026 | ||
| Classification | MHIST binary | 32 | May 19, 2026 | ||
| Image Classification and Calibration | CRC-100K 8-class | 32 | May 19, 2026 | ||
| Process-constrained task execution | FullHome (VirtualHome) | 32 | May 19, 2026 | ||
| Binary Classification | SST-2 (test) | 32 |
| May 19, 2026 |
| Medical Image Segmentation | AMOS MR | 32 | Jun 9, 2026 |
|---|
| Jailbreaking MLLMs | MM-SafetyBench | 32 | May 19, 2026 |
|---|
| Jailbreaking MLLMs | AdvBench | 32 | May 19, 2026 |
|---|
| Multimodal Understanding | MLLM Evaluation Suite (HallusionBench, MME, AI2D, RWQA, SQA, POPE, MMBench, CCB, VSR, V7W) (test) | 32 | May 19, 2026 |
|---|
| Multimodal Understanding | CCB | 32 | May 19, 2026 |
|---|
| Time Series Forecasting | GIFT-Eval 05/18/2025 snapshot (test) | 32 | May 19, 2026 |
|---|
| Adversarial Attack | CIFAR-10 | 32 | May 21, 2026 |
|---|
| Jailbreak Attack | AdvBench 50 harmful behaviors | 32 | May 18, 2026 |
|---|
| Data Reconstruction Attack | ImageNet | 32 | May 18, 2026 |
|---|
| Response Prediction | PDO tumor-selective (aggregated runs) | 32 | May 18, 2026 |
|---|
| Continual Learning | MNIST Random-Label | 32 | May 18, 2026 |
|---|
| Multimodal Contextual Forecasting | Stock Market | 32 | May 15, 2026 |
|---|
| Multimodal Contextual Forecasting | Zillow Real Estate | 32 | May 15, 2026 |
|---|
| Continual Test-Time Adaptation | CCC Hard | 32 | May 15, 2026 |
|---|
| Continual Test-Time Adaptation | CCC Medium | 32 | May 15, 2026 |
|---|
| Continual Test-Time Adaptation | CCC Easy | 32 | May 15, 2026 |
|---|
| Air quality estimation | Real air quality data (test-train) | 32 | May 14, 2026 |
|---|
| JSON Generation | JSON-mode-eval e=1.1 | 32 | May 14, 2026 |
|---|
| Time Series Forecasting | Traffic (10% train) | 32 | May 19, 2026 |
|---|
| Time Series Forecasting | Electricity 10% (train) | 32 | May 19, 2026 |
|---|