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 |
|---|---|---|---|---|---|
| Long-tailed Image Classification | TinyImageNet long-tailed (test) | 30 | May 14, 2026 | ||
| Long-tailed Image Classification | CIFAR100 long-tailed (test) | 30 | May 14, 2026 | ||
| Pruning | MNIST | 30 | May 14, 2026 | ||
| Language Understanding | MMLU | 30 | May 14, 2026 | ||
| Graph Reasoning | OOD graph (test) | 30 | May 14, 2026 | ||
| Reasoning | In-distribution reasoning (test) |
| 30 |
| May 14, 2026 |
| Time Series Forecasting | Weather 5% train data (test) | 30 | May 14, 2026 |
|---|
| Time Series Forecasting | Electricity (5% train) | 30 | May 14, 2026 |
|---|
| Long-form Generation | Alpaca | 30 | May 14, 2026 |
|---|
| Backdoor Defense | LLM Backdoor Defense (test) | 30 | May 14, 2026 |
|---|
| Phenotype Mining | BiolarkGSC+ | 30 | May 14, 2026 |
|---|
| Multimodal Reasoning | SLAKE | 30 | Jul 1, 2026 |
|---|
| Open-domain QA | Overall QA Suite (NQ, TriviaQA, PopQA, HotpotQA, 2WikiMultiHopQA, MuSiQue, Bamboogle) | 30 | May 13, 2026 |
|---|
| Open-domain Single-hop QA | TriviaQA | 30 | May 13, 2026 |
|---|
| PDE Identification | Adv-Diff PDE | 30 | May 13, 2026 |
|---|
| Mathematical Reasoning | Combined Mathematical Reasoning Benchmarks | 30 | May 19, 2026 |
|---|
| General AI Assistant Tasks | GAIA | 30 | May 13, 2026 |
|---|
| Shopping Planning | Shopping Planning | 30 | May 13, 2026 |
|---|
| Travel Planning | Travel Planning | 30 | May 13, 2026 |
|---|
| Binding affinity prediction | SKEMPI v2.0 | 30 | May 13, 2026 |
|---|
| Malicious Training Sample Detection | Attack Training Samples (test) | 30 | May 13, 2026 |
|---|
| Bus Routing Problem | BRP | 30 | May 12, 2026 |
|---|
| Location Routing Problem | LRP | 30 | May 12, 2026 |
|---|
| Traveling Salesman Problem | TSP | 30 | May 12, 2026 |
|---|
| Deepfake Detection | FaceSwapper (test) | 30 | May 12, 2026 |
|---|