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 |
|---|---|---|---|---|---|
| Confidence Calibration | car | 44 | Mar 12, 2026 | ||
| Jailbreaking | MHSC | 44 | Mar 12, 2026 | ||
| Jailbreaking | Q16 | 44 | Mar 12, 2026 | ||
| Mathematical Reasoning | MATH | 44 | May 27, 2026 | ||
| Mathematical Reasoning | MATH 500 | 44 | May 18, 2026 | ||
| Out-of-Distribution Detection |
| CIFAR-100 (In-distribution) vs CIFAR-10 (OOD) (test) |
| 44 |
| Jul 9, 2026 |
| Image Classification | Seq-Tiny-ImageNet | 44 | Mar 10, 2026 |
|---|
| Embodied Navigation | LENL (test) | 44 | Mar 9, 2026 |
|---|
| Dynamic Link Prediction | Wikipedia transductive (ind) | 44 | Jun 4, 2026 |
|---|
| Inductive dynamic link prediction | UN Trade | 44 | Jun 4, 2026 |
|---|
| Inductive dynamic link prediction | Flights | 44 | Jun 4, 2026 |
|---|
| Surface Coverage Path Planning | SPONGE 10 containers | 44 | Mar 4, 2026 |
|---|
| Multi-instruction Visual Reasoning | MM4 | 44 | Mar 4, 2026 |
|---|
| Parallel Code Generation | ParEval 1.0 (full) | 44 | Mar 4, 2026 |
|---|
| Mathematics | Minerva Math | 44 | Jun 2, 2026 |
|---|
| Multiple Instance Learning Classification | RCC | 44 | Mar 4, 2026 |
|---|
| Multiple Instance Learning Classification | Camelyon16 | 44 | Mar 4, 2026 |
|---|
| Offline Reinforcement Learning | HalfCheetah 1T10S (Medium-Replay) | 44 | Mar 4, 2026 |
|---|
| Video Summarization | SumMe (TVT) | 44 | May 12, 2026 |
|---|
| Complex Reasoning | LLaVA Bench (val) | 44 | Mar 4, 2026 |
|---|
| Visual Perception | BLINK (val) | 44 | Jun 2, 2026 |
|---|
| Traveling Salesman Problem | Uniform-TSP1000 | 44 | Jun 9, 2026 |
|---|
| Present Keyphrase Generation | KPTimes (test) | 44 | Feb 27, 2026 |
|---|
| Present Keyphrase Generation | LDKP10K (test) | 44 | Feb 27, 2026 |
|---|
| Present Keyphrase Generation | LDKP3K (test) | 44 | Feb 27, 2026 |
|---|