TreeDebater vs Agent4Debate (Baseline)

Consistent Superior Performance Across Both Model Settings
🏆 TreeDebater outperforms Agent4Debate baseline in both Gemini and DeepSeek settings
Overall Win Rate Comparison - Most Important Result
Gemini Setting
Overall Persuasiveness Across All Stages
Agent4Debate (Baseline)
3.54
TreeDebater (Ours)
3.69
+4.2% improvement in persuasiveness
DeepSeek Setting
Overall Persuasiveness Across All Stages
Agent4Debate (Baseline)
3.47
TreeDebater (Ours)
4.01
+15.6% improvement in persuasiveness
Win Rate by Debate Stage - Gemini Setting
Win Rate by Debate Stage - DeepSeek Setting