Ranking Consensus Aggregation on STRUCTDEBATE Multi-turn
90.1Borda ScoreRAW+SS
Evaluation Results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
| RAW+SS2026.05 | 90.1 | 92 | — | — | — | — | — | |
| RAW2026.05 | 48.7 | 49.2 | — | — | — | — | — | |
| GRASP-W1+D¯2026.05 | 30.8 | 30.9 | — | — | — | — | — | |
| GRASP-W∞+D¯2026.05 | 29.4 | 29.5 | — | — | — | — | — | |
| GRASP-W12026.05 | 28.6 | 28.8 | — | — | — | — | — | |
| GRASP2026.05 | 27.2 | 27.3 | — | — | — | — | — | |
| GRASP-W∞2026.05 | 27.2 | 27.3 | — | — | — | — | — | |
| GRASPnormalization=none, rescale defense matrix=false2026.05 | — | — | 0.626 | 0.187 | 0.779 | 48.7 | 61.9 | |
| GRASP-W∞normalization=L∞, rescale defense matrix=false2026.05 | — | — | 0.626 | 0.187 | 0.779 | 48.6 | 62 | |
| GRASP-W∞+D¯normalization=L∞, rescale defense matrix=true2026.05 | — | — | 0.593 | 0.204 | 0.755 | 52.8 | 63.2 | |
| GRASP-W1normalization=L1, rescale defense matrix=false2026.05 | — | — | 0.607 | 0.197 | 0.768 | 52.5 | 63.4 | |
| GRASP-W1+D¯normalization=L1, rescale defense matrix=true2026.05 | — | — | 0.575 | 0.212 | 0.74 | 54.2 | 63.5 | |
| RAW2026.05 | — | — | 0.309 | 0.345 | 0.38 | 41 | 47.4 | |
| RAW+SSstructural sufficiency guidance=true2026.05 | — | — | 0.064 | 0.468 | 0.038 | 25.1 | 29.8 |