Reasoning quality evaluation on STRATEGYQA
0.2735Somers' DMarODE(αβ)
Evaluation Results
| Method | Links | |
|---|---|---|
| MarODE(αβ)2026.03 | 0.2735 | |
| ReCEval2026.03 | 0.2561 | |
| MarODE2026.03 | 0.2256 | |
| MarODE_QUALITY(β)2026.03 | 0.2213 | |
| MarODE(βγ)2026.03 | 0.1952 | |
| MarODE_COHERENCE(α)2026.03 | 0.1254 | |
| MarODE(αγ)2026.03 | 0.116 | |
| ROSCOE_MEAN2026.03 | 0.0887 | |
| MarODE_EVIDENCE(γ)2026.03 | 0.0744 | |
| ROSCOE-SA2026.03 | 0.0629 | |
| ROSCOE-LI2026.03 | 0.0604 | |
| Local_and_Global_Coherence2026.03 | 0.0489 | |
| ROSCOE-LC2026.03 | 0.0156 | |
| ROSCOE-SS2026.03 | 0.0125 | |
| LLM_as_a_Judge2026.03 | 0.0045 |