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 |
|---|---|---|---|---|---|
| Cancer Subtyping | TCGA-NLCSC (test) | 53 | Feb 26, 2026 | ||
| 3D Dense Captioning | Scan2Cap (val) | 53 | Jun 5, 2026 | ||
| Science Question Answering | SciQA-IMG | 53 | Feb 26, 2026 | ||
| Image Retrieval | NUS-WIDE 5000 | 52 | Jul 7, 2026 | ||
| Text Classification | Edu-Feedback | 52 | Jul 3, 2026 | ||
| Attack Success Rate evaluation |
| Vanilla harmful dataset |
| 52 |
| Jul 3, 2026 |
| Image Classification | ImageNet-1K (val) | 52 | Jul 3, 2026 |
|---|
| Conformalized Regression | Simulation Scenario S1 phi=10 or sigma=0.63 | 52 | Jul 1, 2026 |
|---|
| Brain network diagnosis | REST-meta-MDD site S1 | 52 | Jun 30, 2026 |
|---|
| Evidence retrieval | tau-bench multi-burst-augmented cache | 52 | Jun 30, 2026 |
|---|
| Speed estimation | Naturalistic speed-estimation dataset | 52 | Jun 29, 2026 |
|---|
| Classification | Speed | 52 | Jun 29, 2026 |
|---|
| Concept Removal | Ring-A-Bell Nudity | 52 | Jun 30, 2026 |
|---|
| Class-conditional Image Generation | ImageNet 2009 (val) | 52 | Jun 16, 2026 |
|---|
| Conformal Prediction | 5-class synthetic data (Example 1) Easy Bin (r* > 0.5) 1.0 (test) | 52 | Jun 16, 2026 |
|---|
| Training-based Mimicry Defense | TI-Dataset, DB-Dataset, CelebA-HQ, VGGFace2, WikiArt Average | 52 | Jun 11, 2026 |
|---|
| Automatic Speech Recognition | MyST (test) | 52 | Jun 11, 2026 |
|---|
| Prognostics | XJTU-SY | 52 | Jun 5, 2026 |
|---|
| Text-to-Image Generation | I2P | 52 | Jun 5, 2026 |
|---|
| Image Classification | TinyImageNet | 52 | Jun 4, 2026 |
|---|
| Total Retention | LLaMA-3.1-8B Capability Evaluation Suite | 52 | Jun 4, 2026 |
|---|
| Multimodal Manipulation Detection | NewsCLIPpings (test) | 52 | Jun 4, 2026 |
|---|
| Multimodal Manipulation Detection | MMFakeBench (test) | 52 | Jun 4, 2026 |
|---|
| Multimodal Manipulation Detection | MDSM (test) | 52 | Jun 4, 2026 |
|---|
| Multimodal Manipulation Detection | DGM4 (test) | 52 | Jun 4, 2026 |
|---|