Strategic Planning and Rationalizing on Trees Make LLMs Better Debaters

2025-05-20

AI Summary

Competitive debating is inherently time-constrained, necessitating strategic trade-offs between argument coverage and interactive persuasion—challenges inadequately addressed by existing methods, which struggle to model dynamic argument interactions and adaptive decision-making. This paper introduces TreeDebater, the first framework featuring a dual-tree architecture: a Rehearsal Tree for anticipatory strategic planning and a Debate Flow Tree to explicitly model real-time debate state evolution. The framework integrates time-aware action scheduling, multi-step counterfactual reasoning, iterative revision guided by simulated audience feedback, and multi-agent adversarial simulation. Experimental results demonstrate that TreeDebater significantly outperforms state-of-the-art multi-agent debating systems in both stage-level and full-debate human evaluations. Moreover, its argument selection strategy exhibits substantially higher alignment with expert human debaters’ decision patterns, validating its cognitive fidelity and strategic robustness.

Technology Category

• Multiagent Systems: Agreement, Argumentation & Negotiation
• Knowledge Representation and Reasoning: Argumentation
• Natural Language Processing: Discourse, Pragmatics & Argument Mining

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• User Modeling, Personalization and Recommendation: User modeling and simulation for interactive and conversational systems

Abstract

Winning competitive debates requires sophisticated reasoning and argument skills. There are unique challenges in the competitive debate: (1) The time constraints force debaters to make strategic choices about which points to pursue rather than covering all possible arguments; (2) The persuasiveness of the debate relies on the back-and-forth interaction between arguments, which a single final game status cannot evaluate. To address these challenges, we propose TreeDebater, a novel debate framework that excels in competitive debate. We introduce two tree structures: the Rehearsal Tree and Debate Flow Tree. The Rehearsal Tree anticipates the attack and defenses to evaluate the strength of the claim, while the Debate Flow Tree tracks the debate status to identify the active actions. TreeDebater allocates its time budget among candidate actions and uses the speech time controller and feedback from the simulated audience to revise its statement. The human evaluation on both the stage-level and the debate-level comparison shows that our TreeDebater outperforms the state-of-the-art multi-agent debate system. Further investigation shows that TreeDebater shows better strategies in limiting time to important debate actions, aligning with the strategies of human debate experts.

Problem

• Addresses time constraints in competitive debates by strategic planning
• Enhances persuasiveness through dynamic argument interaction tracking
• Improves debate performance via optimized time allocation and feedback

Innovation

• Uses Rehearsal Tree for claim strength evaluation
• Employs Debate Flow Tree to track debate status
• Integrates speech time controller and audience feedback