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 |
|---|---|---|---|---|---|
| Node Classification | CiteSeer (60%/20%/20%) | 45 | Mar 11, 2026 | ||
| Image Classification | ImageNet (val) | 45 | Feb 26, 2026 | ||
| Unconditional generation | OpenWebText (OWT) L=1024 (held-out) | 45 | Feb 26, 2026 | ||
| Question Answering | TriviaQA (TQA) (test) | 45 | Feb 26, 2026 | ||
| Question Answering | Natural Questions (NQ) (test) | 45 | Feb 26, 2026 | ||
| Table Fact Verification |
|---|
| InfoTabs |
| 45 |
| Apr 7, 2026 |
| Table Question Answering | TAT-QA | 45 | Apr 7, 2026 |
|---|
| Entity Regression | RelBench v1.0 (test) | 45 | May 11, 2026 |
|---|
| Graph Generation | Enzymes | 45 | Apr 3, 2026 |
|---|
| Mathematical Reasoning | GSM8K | 45 | Feb 26, 2026 |
|---|
| Question Answering | DisambiguationQA | 45 | Feb 26, 2026 |
|---|
| Audio Assessment Correlation | PAM | 45 | Jun 4, 2026 |
|---|
| Traveling Salesman Problem | TSP 1K (test) | 45 | Apr 16, 2026 |
|---|
| Question Answering | LoCoMo | 45 | Apr 21, 2026 |
|---|
| Question Answering | DROP | 45 | May 21, 2026 |
|---|
| Molecular Generation | ZINC 250k | 45 | May 19, 2026 |
|---|
| Code Generation | LiveCodeBench v6 | 45 | Jul 2, 2026 |
|---|
| Function Selection | TaskBench HuggingFace | 45 | Feb 26, 2026 |
|---|
| Capacitated Vehicle Routing Problem | CVRPLib (test) | 45 | Jun 11, 2026 |
|---|
| Time Series Forecasting | EURUSD (test) | 45 | Feb 26, 2026 |
|---|
| Time Series Forecasting | WTI (test) | 45 | Feb 26, 2026 |
|---|
| Time Series Forecasting | S&P500 (test) | 45 | Feb 26, 2026 |
|---|
| Node Classification | CiteSeer | 45 | Mar 23, 2026 |
|---|
| Mathematical Reasoning | AIME25 (test) | 45 | Jun 16, 2026 |
|---|
| Adjacency Matrix Reconstruction | Graph Data Instances | 45 | Feb 26, 2026 |
|---|