Question Answering on MMLU (Acc, Δ, Len, Δ%)
65.27AccuracyOriginal Model
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| Original ModelBackbone=DeepSeek-R1-Distill-Qwen-7B2025.08 | 65.27 | — | 1,858 | — | |
| LCPOBackbone=DeepSeek-R1-Distill-Qwen-7B2025.08 | 64.16 | -1.11 | 835 | -55.06 | |
| LCPOBackbone=DeepSeek-R1-Distill-Qwen-1.5B2025.08 | 46.78 | 0.35 | 760 | -70.18 | |
| Original ModelBackbone=DeepSeek-R1-Distill-Qwen-1.5B2025.08 | 46.43 | — | 2,549 | — |