General Performance on Overall Aggregate (test)
72.4Average ScoreFull
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| FullModel=Qwen3, Pruning Ratio=0%, Criterion=Full, Calibration=C42026.06 | 72.4 | — | |
| MANModel=Qwen3, Pruning Ratio=25%, Criterion=MAN, (b, α, β)=(1, 0, 1), Calibration=C42026.06 | 70.7 | 2.75 | |
| MSANModel=Qwen3, Pruning Ratio=25%, Criterion=MSAN, (b, α, β)=(1, 0, 2), Calibration=C42026.06 | 70 | 3.25 | |
| FullModel=ERNIE, Pruning Ratio=0%, Criterion=Full, Calibration=C42026.06 | 65.4 | — | |
| REAPModel=Qwen3, Pruning Ratio=25%, Criterion=REAP, (b, α, β)=(1, 1, 1), Calibration=C42026.06 | 61.9 | 4.67 | |
| MANModel=ERNIE, Pruning Ratio=25%, Criterion=MAN, (b, α, β)=(1, 0, 1), Calibration=C42026.06 | 48.5 | 2 | |
| MSANModel=ERNIE, Pruning Ratio=25%, Criterion=MSAN, (b, α, β)=(1, 0, 2), Calibration=C42026.06 | 48.4 | 2 | |
| REAPModel=ERNIE, Pruning Ratio=25%, Criterion=REAP, (b, α, β)=(1, 1, 1), Calibration=C42026.06 | 45.6 | 3.25 | |
| FullModel=DeepSeek, Pruning Ratio=0%, Criterion=Full, Calibration=C42026.06 | 44.3 | — | |
| MoNEModel=ERNIE, Pruning Ratio=25%, Criterion=MoNE, Calibration=C42026.06 | 42.1 | 4.33 | |
| EANModel=ERNIE, Pruning Ratio=25%, Criterion=EAN, (b, α, β)=(0, 0, 1), Calibration=C42026.06 | 41.5 | 4.5 | |
| MoNEModel=Qwen3, Pruning Ratio=25%, Criterion=MoNE, Calibration=C42026.06 | 41.3 | 4.17 | |
| SEERModel=ERNIE, Pruning Ratio=25%, Criterion=SEER, (b, α, β)=(0, 1, 0), Calibration=C42026.06 | 41.2 | 5.25 | |
| FrequencyModel=Qwen3, Pruning Ratio=25%, Criterion=Frequency, (b, α, β)=(0, 0, 0), Calibration=C42026.06 | 41.1 | 4.83 | |
| SEERModel=Qwen3, Pruning Ratio=25%, Criterion=SEER, (b, α, β)=(0, 1, 0), Calibration=C42026.06 | 41.1 | 3.92 | |
| EANModel=Qwen3, Pruning Ratio=25%, Criterion=EAN, (b, α, β)=(0, 0, 1), Calibration=C42026.06 | 41.1 | 4.42 | |
| FullModel=OLMoE, Pruning Ratio=0%, Criterion=Full, Calibration=C42026.06 | 39.6 | — | |
| FrequencyModel=ERNIE, Pruning Ratio=25%, Criterion=Frequency, (b, α, β)=(0, 0, 0), Calibration=C42026.06 | 38 | 6.67 | |
| MSANModel=DeepSeek, Pruning Ratio=25%, Criterion=MSAN, (b, α, β)=(1, 0, 2), Calibration=C42026.06 | 23.7 | 2 | |
| EANModel=DeepSeek, Pruning Ratio=25%, Criterion=EAN, (b, α, β)=(0, 0, 1), Calibration=C42026.06 | 20.9 | 3.33 | |
| MANModel=OLMoE, Pruning Ratio=25%, Criterion=MAN, (b, α, β)=(1, 0, 1), Calibration=C42026.06 | 18.6 | 2 | |
| REAPModel=DeepSeek, Pruning Ratio=25%, Criterion=REAP, (b, α, β)=(1, 1, 1), Calibration=C42026.06 | 18.5 | 3.92 | |
| MANModel=DeepSeek, Pruning Ratio=25%, Criterion=MAN, (b, α, β)=(1, 0, 1), Calibration=C42026.06 | 18.5 | 3.33 | |
| MSANModel=OLMoE, Pruning Ratio=25%, Criterion=MSAN, (b, α, β)=(1, 0, 2), Calibration=C42026.06 | 18.1 | 2.5 | |
| MoNEModel=DeepSeek, Pruning Ratio=25%, Criterion=MoNE, Calibration=C42026.06 | 18 | 4.42 | |
| REAPModel=OLMoE, Pruning Ratio=25%, Criterion=REAP, (b, α, β)=(1, 1, 1), Calibration=C42026.06 | 17.3 | 2.83 | |
| FrequencyModel=DeepSeek, Pruning Ratio=25%, Criterion=Frequency, (b, α, β)=(0, 0, 0), Calibration=C42026.06 | 15.5 | 5.33 | |
| EANModel=OLMoE, Pruning Ratio=25%, Criterion=EAN, (b, α, β)=(0, 0, 1), Calibration=C42026.06 | 15.2 | 4.58 | |
| MoNEModel=OLMoE, Pruning Ratio=25%, Criterion=MoNE, Calibration=C42026.06 | 14.8 | 5 | |
| SEERModel=DeepSeek, Pruning Ratio=25%, Criterion=SEER, (b, α, β)=(0, 1, 0), Calibration=C42026.06 | 13.4 | 5.67 | |
| SEERModel=OLMoE, Pruning Ratio=25%, Criterion=SEER, (b, α, β)=(0, 1, 0), Calibration=C42026.06 | 12.6 | 5.42 | |
| FrequencyModel=OLMoE, Pruning Ratio=25%, Criterion=Frequency, (b, α, β)=(0, 0, 0), Calibration=C42026.06 | 12.4 | 5.67 |