Contrastive Summarization on TravelDest
0.75ContrastQ-STRUM Debate
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| Q-STRUM DebateComparison Baseline=Q-STRUM Base, Evaluator Model=Claude-3.5-Sonnet2025.02 | 0.75 | 0.58 | 0.53 | 0.63 | |
| Q-STRUM DebateComparison Baseline=Q-STRUM Contrastive, Evaluator Model=Claude-3.5-Sonnet2025.02 | 0.7 | 0.5 | 0.48 | 0.54 |