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 |
|---|---|---|---|---|---|
| Long-context Retrieval | MLDR | 36 | Jun 23, 2026 | ||
| Image Classification | CIFAR10 | 36 | Jun 11, 2026 | ||
| Gold-standard ranking agreement | combined benchmark | 36 | Mar 12, 2026 | ||
| Spatio-temporal forecasting | GLA | 36 | Mar 12, 2026 | ||
| Class-Incremental Learning | CUB (test) | 36 | Mar 12, 2026 | ||
| Image Classification |
| MTIL task-agnostic (test) |
| 36 |
| Jun 16, 2026 |
| Mathematical Reasoning | AMC 2023 | 36 | Jun 17, 2026 |
|---|
| Tracking | ScanNet | 36 | Jun 30, 2026 |
|---|
| Speaker Verification | VoxCeleb1 with Nonspeech100 (test) | 36 | Mar 11, 2026 |
|---|
| Semi-Supervised Domain Generalization Image Classification | Retina (test) | 36 | Mar 10, 2026 |
|---|
| Semi-Supervised Domain Generalization Image Classification | ESCA (test) | 36 | Mar 10, 2026 |
|---|
| Pathological speech intelligibility assessment | COPAS Sentence | 36 | Mar 10, 2026 |
|---|
| Pathological speech intelligibility assessment | COPAS Word | 36 | Mar 10, 2026 |
|---|
| Node Classification Robustness Certification | Wisconsin | 36 | Mar 10, 2026 |
|---|
| Node Classification Robustness Certification | Cornell | 36 | Mar 10, 2026 |
|---|
| Image Denoising | Set14 (test) | 36 | Apr 15, 2026 |
|---|
| Multiresolution Rendering | TUM dataset | 36 | Mar 10, 2026 |
|---|
| Machine Unlearning | CIFAR-100 (forget set) | 36 | Mar 10, 2026 |
|---|
| End-to-end Time-to-First-Token (TTFT) | Long-context sequences | 36 | Mar 9, 2026 |
|---|
| Time Series Forecasting | EPF NP | 36 | Jun 11, 2026 |
|---|
| Preference Alignment | TL;DR (test) | 36 | Mar 6, 2026 |
|---|
| Preference Alignment | HH-RLHF (test) | 36 | Jun 1, 2026 |
|---|
| Untargeted Adversarial Attack | ImageNet | 36 | Mar 6, 2026 |
|---|
| Subword-to-byte distribution approximation | reference distribution token-to-byte 2025a (test) | 36 | Mar 6, 2026 |
|---|
| Sequence Scoring | Held-out text (G.4) | 36 | Mar 6, 2026 |
|---|