Reasoning on GSM8K (Somers' D)
0.1858Somers' DMarODE
Evaluation Results
| Method | Links | |
|---|---|---|
| MarODE2026.03 | 0.1858 | |
| MarODE_COHERENCE(α)2026.03 | 0.1809 | |
| ReCEval2026.03 | 0.0505 | |
| MarODE_QUALITY(β)2026.03 | 0.0212 | |
| Local_and_Global_Coherence2026.03 | 0.0165 | |
| LLM_as_a_Judge2026.03 | 0.0028 | |
| ROSCOE-LI2026.03 | -0.022 | |
| ROSCOE-SS2026.03 | -0.0325 | |
| ROSCOE_MEAN2026.03 | -0.0325 | |
| ROSCOE-SA2026.03 | -0.0443 | |
| ROSCOE-LC2026.03 | -0.0536 |