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 |
|---|---|---|---|---|---|
| Text classification | Drug Review Norethindrone (5-fold cross-validation) | 36 | Feb 26, 2026 | ||
| Text classification | Newsgroup Religion (5-fold cross-validation) | 36 | Feb 26, 2026 | ||
| Text classification | Newsgroup Science (5-fold cross-validation) | 36 | Feb 26, 2026 | ||
| LLM-as-a-judge evaluation | MT-Bench | 36 | Mar 19, 2026 | ||
| LLM-as-a-judge evaluation | FLASK | 36 |
| Mar 19, 2026 |
| LLM-as-a-judge evaluation | FB Bench (Feedback Bench) | 36 | Mar 19, 2026 |
|---|
| Visual Understanding | SEED-Bench | 36 | Jun 23, 2026 |
|---|
| Contrastive Alignment | Must-C (test) | 36 | Feb 26, 2026 |
|---|
| Joint Word Segmentation and POS Tagging | Japanese (test) | 36 | Feb 26, 2026 |
|---|
| Joint Word Segmentation and POS Tagging | Chinese (test) | 36 | Feb 26, 2026 |
|---|
| Multi-agent Scenario Generation | nuScenes (train) | 36 | Feb 26, 2026 |
|---|
| General Reasoning | BIG-bench | 36 | Apr 14, 2026 |
|---|
| 3D Classification | ScanObjectNN OBJ_BG (test) | 36 | Feb 26, 2026 |
|---|
| Base-to-novel generalization | UCF101 base-to-novel | 36 | Jul 2, 2026 |
|---|
| Multimodal reasoning | MathVista | 36 | Mar 24, 2026 |
|---|
| Medical Image Classification | MedMNIST Breast (test) | 36 | Feb 26, 2026 |
|---|
| Medical Image Classification | MedMNIST Derma (test) | 36 | Feb 26, 2026 |
|---|
| Medical Image Classification | MedMNIST Pneumonia (test) | 36 | Feb 26, 2026 |
|---|
| Sleep Stage Scoring | Sleep (test) | 36 | Feb 26, 2026 |
|---|
| Time Series Classification | fNIRS (test) | 36 | Feb 26, 2026 |
|---|
| Machine Translation | De-En document-level | 36 | Feb 26, 2026 |
|---|
| 3D Object Detection | View-of-Delft (VoD) (val) | 36 | May 21, 2026 |
|---|
| Image-to-Text Retrieval | Urban1k | 36 | May 13, 2026 |
|---|
| Hallucination Evaluation | MSCOCO (val) | 36 | Feb 26, 2026 |
|---|
| Hand Pose Estimation | DexYCB (S1) | 36 | Feb 26, 2026 |
|---|