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 |
|---|---|---|---|---|---|
| Mathematical Problem Solving | AIME 2024 | 54 | May 18, 2026 | ||
| Language Generation | Wiki | 54 | Mar 27, 2026 | ||
| Traffic state diagram refinement | I-24 MOTION INCEPTION VT v1.0 (test Nov 30, Dec 1, 2) | 54 | Mar 26, 2026 | ||
| Multi-turn Visual Question Answering | ConvBench | 54 | Mar 24, 2026 | ||
| Medical Question Answering | MedMCQA |
| 54 |
| Mar 19, 2026 |
| Jailbreak Defense | JBC | 54 | Mar 19, 2026 |
|---|
| Mathematical Reasoning | OlymBench | 54 | Jun 23, 2026 |
|---|
| Conditional Average Treatment Effect Estimation | Simulation study n1=250 | 54 | Mar 19, 2026 |
|---|
| 3D Semantic Occupancy Prediction | Occ3D-nuScenes v1.0 (val) | 54 | Jul 1, 2026 |
|---|
| Language Model Unlearning | TOFU (Forget10) | 54 | May 4, 2026 |
|---|
| Node Classification | Computers-Random (ρ ≈ 17.7) | 54 | Mar 18, 2026 |
|---|
| Node Classification | CS ρ ≈ 41.0 (random) | 54 | Mar 18, 2026 |
|---|
| Machine Unlearning | TOFU Forget 1% | 54 | May 14, 2026 |
|---|
| Large-scale Sparse Multi-objective Optimization | SMOP2 | 54 | Jul 7, 2026 |
|---|
| Large-scale Sparse Multi-objective Optimization | SMOP 1 | 54 | Jul 7, 2026 |
|---|
| Multi-hop Question Answering | HotPotQA | 54 | Mar 13, 2026 |
|---|
| General QA | NQ | 54 | May 11, 2026 |
|---|
| Spike Sequence Classification | NeuroMorse | 54 | Mar 11, 2026 |
|---|
| Cross-modal Geo-localization | CVG-Text (New York) | 54 | Jun 23, 2026 |
|---|
| Appearance Reconstruction | Waymo 8 scenes | 54 | Mar 10, 2026 |
|---|
| Dataset-level accuracy estimation | WikiSQL to Spider 2.0 | 54 | Mar 10, 2026 |
|---|
| Dataset-level accuracy estimation | Spider to SynSQL 2.5M | 54 | Mar 10, 2026 |
|---|
| Dataset-level accuracy estimation | SParC to CoSQL | 54 | Mar 10, 2026 |
|---|
| Dataset-level accuracy estimation | WikiSQL to Spider | 54 | Mar 10, 2026 |
|---|
| Dataset-level accuracy estimation | Spider to BIRD | 54 | Mar 10, 2026 |
|---|