SOTA2 Research · benchmarks
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
Benchmarks
309,131 indexed| Task | Dataset | Domain | Trend | Results | Last Update |
|---|---|---|---|---|---|
| Initial Structure to Relaxed Energy (IS2RE) | OC20 (Open Catalyst 2020) IS2RE (test) | 30 | Feb 26, 2026 | ||
| S2EF (Structure to Energy and Forces) | OC20 average across all four splits (val) | 30 | Feb 26, 2026 | ||
| Molecular Property Prediction (Regression) | MoleculeNet (test) | 30 | Jun 11, 2026 | ||
| top tagging | Top Tagging Benchmark Dataset | 30 | Feb 26, 2026 | ||
| Image Compression | CelebA (test) | 30 | May 1, 2026 | ||
| Speech Enhancement | DNS no-reverb 2020 (test) | 30 | Jul 9, 2026 | ||
| Class conditional image generation | ImageNet 64x64 (train test) | 30 | Feb 26, 2026 | ||
| Person Re-Identification | MSMT17 -> CUHK03-NP (test) | 30 | Mar 17, 2026 | ||
| Red blood cell classification | Elsafty (leave-one-source-out) | 30 | Feb 26, 2026 | ||
| Hierarchical Table Question Answering | MultiHiertt (test) | 30 | Feb 26, 2026 | ||
| Conversational Financial Question Answering | ConvFinQA (test) | 30 | Feb 26, 2026 | ||
| Financial Question Answering | TATQA 1.0 (test) | 30 | Feb 26, 2026 | ||
| Co-Speech Gesture Video Generation | PATS (test) | 30 | Jun 4, 2026 | ||
| Meme Classification | MAMI | 30 | Feb 26, 2026 | ||
| Meme Classification | HarMeme | 30 | Feb 26, 2026 | ||
| Semantic Segmentation | MFNet daytime (test) | 30 | Feb 26, 2026 | ||
| Instance Segmentation | ADE20K 150 59 (val) | 30 | Feb 26, 2026 | ||
| Data Imputation | Gliomas | 30 | Feb 26, 2026 | ||
| Data Imputation | NPHA | 30 | Feb 26, 2026 | ||
| Anomaly Detection | OCT 2017 | 30 | May 28, 2026 | ||
| Vulnerability Detection | Bigvul (test) | 30 | Feb 26, 2026 | ||
| Code Debugging | MBPP | 30 | Feb 26, 2026 | ||
| Document Visual Question Answering | DUDE | 30 | Feb 26, 2026 | ||
| Defense against Indirect Prompt Injection | Filtered QA dataset | 30 | Feb 26, 2026 | ||
| Direct Prompt Injection | AlpacaFarm (208 samples) | 30 | Feb 26, 2026 |