多智能体辩论提升论点分类准确率，无需领域训练。
Multi-Agent Dialectical Refinement for Enhanced Argument Classification
- 用正方反方裁判三智能体辩论解析模糊论点结构
- 在学生作文数据集上达85.7%宏平均F1，超越单模型基线
- 输出可读辩论记录，结果透明可解释，适合教育评估场景
论点挖掘（AM）是自动化写作评价的基础技术，但传统监督方法依赖昂贵的领域特定微调。尽管大语言模型（LLMs）提供无训练替代方案，却常因结构歧义难以区分主张与前提等相似组件。此外，单智能体自我修正机制易陷入自我说服，强化初始错误。本文提出MAD-ACC（多智能体辩论论点成分分类框架），通过辩证精炼解决分类不确定性。该框架采用正方-反方-裁判模型，让智能体对模糊文本的不同解释进行辩护，揭示单模型遗漏的逻辑细节。在UKP学生作文语料库上的评估显示，MAD-ACC实现85.7%的宏平均F1，显著优于单智能体推理基线，且无需领域特定训练。同时，不同于“黑箱”分类器，其辩证方法生成人类可读的辩论记录，清晰展现决策依据，提供透明可解释的替代方案。
原文摘要 · Abstract (English)
Argument Mining (AM) is a foundational technology for automated writing evaluation, yet traditional supervised approaches rely heavily on expensive, domain-specific fine-tuning. While Large Language Models (LLMs) offer a training-free alternative, they often struggle with structural ambiguity, failing to distinguish between similar components like Claims and Premises. Furthermore, single-agent self-correction mechanisms often suffer from sycophancy, where the model reinforces its own initial errors rather than critically evaluating them. We introduce MAD-ACC (Multi-Agent Debate for Argument Component Classification), a framework that leverages dialectical refinement to resolve classification uncertainty. MAD-ACC utilizes a Proponent-Opponent-Judge model where agents defend conflicting interpretations of ambiguous text, exposing logical nuances that single-agent models miss. Evaluation on the UKP Student Essays corpus demonstrates that MAD-ACC achieves a Macro F1 score of 85.7%, significantly outperforming single-agent reasoning baselines, without requiring domain-specific training. Additionally, unlike "black-box" classifiers, MAD-ACC's dialectical approach offers a transparent and explainable alternative by generating human-readable debate transcripts that explain the reasoning behind decisions.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。