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 |
|---|---|---|---|---|---|
| Helpfulness alignment | HHH alignment | 44 | May 19, 2026 | ||
| DeepFake Detection | DF40 (FR) | 44 | Jun 2, 2026 | ||
| DeepFake Detection | DF40 FS | 44 | Jun 2, 2026 | ||
| Commonsense Reasoning | Commonsense Reasoning Suite Boolq, PIQA, SIQA, Win, OBQA, HellaSwag, ARC-E, ARC-C | 44 | May 22, 2026 | ||
| Multi-Hop Question Answering | Bamboogle |
| 44 |
| May 14, 2026 |
| Multi-Hop Question Answering | 2Wiki | 44 | May 14, 2026 |
|---|
| Model Merging | 20-task vision merging scenario (test) | 44 | May 14, 2026 |
|---|
| Model Merging | 14-task vision merging scenario (test) | 44 | May 14, 2026 |
|---|
| Model Merging | Language Benchmarks 5-task | 44 | May 14, 2026 |
|---|
| Hidden objective detection | Prompt-Manipulation Out-of-Domain Evaluation Stealth Strategy | 44 | May 14, 2026 |
|---|
| Hidden Objective Detection | Hidden Objective Detection Evaluation Set Dependency 1.0 (out-of-domain) | 44 | May 14, 2026 |
|---|
| Image Classification | ImageNet-1K (val) | 44 | May 13, 2026 |
|---|
| Memory Agent Performance | MemoryAgentBench | 44 | Jun 11, 2026 |
|---|
| Image-to-Text Retrieval | Flickr30K | 44 | Jun 30, 2026 |
|---|
| Multi-agent consensus coordination | Multi-Agent LLM Consensus System 50 samples per configuration | 44 | May 12, 2026 |
|---|
| Visual Quality Evaluation | EBench-18K | 44 | May 11, 2026 |
|---|
| Multi-turn Dialogue | MT-Bench | 44 | May 29, 2026 |
|---|
| Commonsense Reasoning | HellaS | 44 | May 12, 2026 |
|---|
| Mathematical Reasoning | Mathematics Benchmarks Average | 44 | May 22, 2026 |
|---|
| Active Modality Acquisition | MIMIC Symile G_full (test) | 44 | May 8, 2026 |
|---|
| Safety Evaluation | XSTest | 44 | May 7, 2026 |
|---|
| Classification | Higgs 30K subsamples | 44 | May 6, 2026 |
|---|
| Classification | Poker Hand | 44 | May 6, 2026 |
|---|
| Classification | Covertype 30K subsamples | 44 | May 6, 2026 |
|---|
| Classification | MiniBooNE 30K subsamples | 44 | May 6, 2026 |
|---|