Knowledge Evaluation on MMLU-Redux 2.0 (Continual)
33.49AccuracySTOC
Evaluation Results
| Method | Links | |
|---|---|---|
| STOCModel Scale=1.7B, Prompts=5-shot, Freezing Layers=Best reported2026.05 | 33.49 | |
| STOCModel Scale=0.6B, Prompts=5-shot, Freezing Layers=Best reported2026.05 | 32.93 | |
| LAMOLModel Scale=1.7B, Prompts=5-shot, Freezing Layers=Best reported2026.05 | 31.71 | |
| LAMOLModel Scale=0.6B, Prompts=5-shot, Freezing Layers=Best reported2026.05 | 28.05 | |
| NaiveModel Scale=0.6B, Prompts=5-shot, Freezing Layers=Best reported2026.05 | 24.39 | |
| NaiveModel Scale=1.7B, Prompts=5-shot, Freezing Layers=Best reported2026.05 | 24.17 |