| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| LLaMA Original Languages 3.1-8B | LLaMA-3.1-8B-Instruct | Generalization Score80.37 | 8 | 15d ago | |
| LLaMA Evaluation Expanded Languages 3.1-8B | DeltaMoE | Overall Score69.37 | 8 | 15d ago | |
| Gen | Modular Gradient Surgery | Gen Score35.7 | 8 | 3mo ago | |
| Single scenario unseen tasks | FedRouter | ROUGE-158.3 | 6 | 2mo ago | |
| Synthetic Dataset | AdverISF (Two-Stage) | R^2 (10%)0.251 | 5 | 3mo ago |