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 |
|---|---|---|---|---|---|
| Out-of-Distribution Detection | CIFAR-10 (test) | 52 | May 8, 2026 | ||
| Event Video Retrieval | EVVE | 52 | Feb 26, 2026 | ||
| Joint Entity and Relation Extraction | WebNLG (test) | 52 | Feb 26, 2026 | ||
| Optic Flow Estimation | MVSEC (outdoor_day1) | 52 | May 26, 2026 | ||
| Classification | LETTER (test) |
| 52 |
| May 11, 2026 |
| Named Entity Recognition | Resume | 52 | Feb 26, 2026 |
|---|
| Salient Object Detection | NLPR | 52 | Apr 21, 2026 |
|---|
| Salient Object Detection | NJUD | 52 | Apr 21, 2026 |
|---|
| Node Classification | Pubmed full-supervised | 52 | Jun 11, 2026 |
|---|
| Nested Named Entity Recognition | ACE 2005 | 52 | Feb 26, 2026 |
|---|
| Speech Recognition | WSJ nov93 (dev) | 52 | Feb 26, 2026 |
|---|
| Depth Prediction | Cityscapes (test) | 52 | May 6, 2026 |
|---|
| Retinal Vessel Segmentation | DRIVE (test) | 52 | Apr 16, 2026 |
|---|
| Partial Domain Adaptation | Office-31 | 52 | Feb 26, 2026 |
|---|
| Vision-and-Language Navigation | Room-to-Room (R2R) Unseen (val) | 52 | Feb 26, 2026 |
|---|
| Phoneme Recognition | TIMIT (test) | 52 | Jun 8, 2026 |
|---|
| Out-of-distribution Detection | CIFAR-100 in-distribution TinyImageNet out-of-distribution (test) | 52 | Jul 9, 2026 |
|---|
| Variational Inference | MNIST (test) | 52 | Mar 10, 2026 |
|---|
| Domain Adaptation | VisDA 2017 (val) | 52 | Feb 26, 2026 |
|---|
| Fine-Grained Visual Classification | Stanford Dogs (test) | 52 | Feb 26, 2026 |
|---|
| Machine Translation (Chinese-to-English) | NIST 2003 (MT-03) | 52 | Feb 26, 2026 |
|---|
| Abstractive Summarization | CNN/Daily Mail non-anonymous (test) | 52 | Feb 26, 2026 |
|---|
| Generalized Zero-Shot Learning | FLO | 52 | Mar 19, 2026 |
|---|
| Activity Detection | Charades localize v1 | 52 | Feb 26, 2026 |
|---|
| Object Detection | ILSVRC weakly labeled categories 2013 (val2) | 52 | Feb 26, 2026 |
|---|