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 |
|---|---|---|---|---|---|
| Instruction-based Image Editing | ImgEdit Bench 1.0 (test) | 37 | Feb 26, 2026 | ||
| Vision-and-Language Navigation | OpenUAV Seen Hard (test) | 37 | Jul 1, 2026 | ||
| Vision-and-Language Navigation | OpenUAV Easy (test seen) | 37 | Jul 1, 2026 | ||
| Vision-Language Navigation | OpenUAV test unseen 1.0 (full) | 37 | Mar 24, 2026 | ||
| GUI Grounding | OSWorld G-Refine v1.0 (test) |
| 37 |
| Jun 15, 2026 |
| Chart Understanding and Reasoning | CharXiv | 37 | Jun 2, 2026 |
|---|
| Infrared Small Target Detection | NUAA-SIRST | 37 | Mar 27, 2026 |
|---|
| Vision-and-Language Navigation | GSA-R2R N-Scene (test) | 37 | May 13, 2026 |
|---|
| Pulse Rate Estimation | PURE to MMPD (test) | 37 | Jun 23, 2026 |
|---|
| Streaming Video Understanding | RVS-Ego | 37 | Jun 19, 2026 |
|---|
| Hallucination Detection | SLAKE (All) | 37 | May 21, 2026 |
|---|
| Hallucination Detection | SLAKE Open-Ended | 37 | May 21, 2026 |
|---|
| Hallucination Detection | VQA-Med Open-Ended 2019 | 37 | May 21, 2026 |
|---|
| Adversarial Attack | MVBench | 37 | Feb 26, 2026 |
|---|
| Adversarial Attack | Q-Bench | 37 | Feb 26, 2026 |
|---|
| Adversarial Attack | BLINK | 37 | Feb 26, 2026 |
|---|
| Adversarial Attack | NLVR2 | 37 | Feb 26, 2026 |
|---|
| Adversarial Attack | Mantis Eval | 37 | Feb 26, 2026 |
|---|
| Camouflaged Object Detection | CAMO | 37 | Mar 26, 2026 |
|---|
| Image Denoising | FMD (test) | 37 | May 26, 2026 |
|---|
| Sign prediction | Wiki-RfA | 37 | Feb 26, 2026 |
|---|
| Membership Inference Attack | CIFAR100 (test) | 37 | Feb 26, 2026 |
|---|
| Visual Question Answering | VQA v2 | 37 | Feb 26, 2026 |
|---|
| Visual Question Answering | TextVQA | 37 | Feb 26, 2026 |
|---|
| Action Recognition | DVS-Gesture | 37 | Jun 4, 2026 |
|---|