Formal Logic Reasoning on MMLU (test)
66.6Accuracy (MMLU FLR Test)ReM-MoA*
Evaluation Results
| Method | Links | |
|---|---|---|
| ReM-MoA*Width (N)=8, Depth (L)=3, Trend=↑↑, Distilled Reviewer=true2026.06 | 66.6 | |
| ReM-MoA*Width (N)=7, Depth (L)=3, Trend=↑↑, Distilled Reviewer=true2026.06 | 65.8 | |
| ReM-MoA*Width (N)=6, Depth (L)=3, Trend=↑↑, Distilled Reviewer=true2026.06 | 64.9 | |
| ReM-MoAWidth (N)=8, Depth (L)=3, Trend=↑↑2026.06 | 64.5 | |
| ReM-MoAWidth (N)=7, Depth (L)=3, Trend=↑↑2026.06 | 63.8 | |
| ReM-MoA*Width (N)=5, Depth (L)=3, Trend=↑↑, Distilled Reviewer=true2026.06 | 63.7 | |
| ReM-MoAWidth (N)=6, Depth (L)=3, Trend=↑↑2026.06 | 63 | |
| ReM-MoA*Width (N)=4, Depth (L)=3, Trend=↑↑, Distilled Reviewer=true2026.06 | 62.3 | |
| ReM-MoAWidth (N)=5, Depth (L)=3, Trend=↑↑2026.06 | 61.8 | |
| ReM-MoA*Width (N)=3, Depth (L)=3, Trend=↑↑, Distilled Reviewer=true2026.06 | 60.8 | |
| ReM-MoAWidth (N)=4, Depth (L)=3, Trend=↑↑2026.06 | 60.5 | |
| AttentionMoAWidth (N)=8, Depth (L)=3, Trend=↑↑2026.06 | 59 | |
| ReM-MoAWidth (N)=3, Depth (L)=3, Trend=↑↑2026.06 | 59 | |
| AttentionMoAWidth (N)=7, Depth (L)=3, Trend=↑↑2026.06 | 58.6 | |
| AttentionMoAWidth (N)=6, Depth (L)=3, Trend=↑↑2026.06 | 58.2 | |
| AttentionMoAWidth (N)=5, Depth (L)=3, Trend=↑↑2026.06 | 57.5 | |
| AttentionMoAWidth (N)=4, Depth (L)=3, Trend=↑↑2026.06 | 56.8 | |
| AttentionMoAWidth (N)=3, Depth (L)=3, Trend=↑↑2026.06 | 55.6 | |
| RMoAWidth (N)=8, Depth (L)=3, Trend=↑→2026.06 | 55.3 | |
| RMoAWidth (N)=7, Depth (L)=3, Trend=↑→2026.06 | 55.2 | |
| RMoAWidth (N)=6, Depth (L)=3, Trend=↑→2026.06 | 55 | |
| RMoAWidth (N)=5, Depth (L)=3, Trend=↑→2026.06 | 54.5 | |
| RMoAWidth (N)=4, Depth (L)=3, Trend=↑→2026.06 | 53.8 | |
| RMoAWidth (N)=3, Depth (L)=3, Trend=↑→2026.06 | 52.8 | |
| Standard MoAWidth (N)=6, Depth (L)=3, Trend=→2026.06 | 52.5 | |
| Standard MoAWidth (N)=7, Depth (L)=3, Trend=→2026.06 | 52.4 | |
| Standard MoAWidth (N)=5, Depth (L)=3, Trend=→2026.06 | 52.2 | |
| Standard MoAWidth (N)=8, Depth (L)=3, Trend=→2026.06 | 52.2 | |
| Standard MoAWidth (N)=4, Depth (L)=3, Trend=→2026.06 | 51.8 | |
| Standard MoAWidth (N)=3, Depth (L)=3, Trend=→2026.06 | 51 |