Language Understanding on MMLU (Accuracy and Average token length)
58.3AccuracyDAST
Evaluation results
| Method | Links | ||
|---|---|---|---|
| DASTBase Model=DeepSeek-R1-Distill-Qwen-7B2026.05 | 58.3 | 1,612 | |
| SLATBase Model=DeepSeek-R1-Distill-Qwen-7B2026.05 | 58.3 | 756 | |
| LC-R1Base Model=DeepSeek-R1-Distill-Qwen-7B2026.05 | 58 | 880 | |
| L1-MaxBase Model=DeepSeek-R1-Distill-Qwen-7B2026.05 | 55.3 | 973 | |
| DeepSeek-R1-Distill-Qwen-7Bstatus=Original base model2026.05 | 53.2 | 1,674 | |
| TLMREBase Model=DeepSeek-R1-Distill-Qwen-7B2026.05 | 50.8 | 1,023 |