多智能体辩证细化用于增强论证分类
Multi-Agent Dialectical Refinement for Enhanced Argument Classification
AI总结：
本文提出MAD-ACC框架，通过多智能体辩论解决论证分类不确定性，实现85.7%的宏F1分数，无需领域特定训练，提供可解释的辩论记录。
摘要：
论证挖掘（AM）是自动写作评估的基础技术，但传统监督方法依赖昂贵的领域特定微调。尽管大型语言模型（LLMs）提供无训练替代方案，但它们在结构模糊性上表现不佳，难以区分相似组件如主张和前提。此外，单智能体自我纠正机制常因讨好而强化初始错误。我们引入MAD-ACC（多智能体辩论用于论证组件分类），通过辩证细化解决分类不确定性。MAD-ACC采用主张-反对-裁判模型，其中智能体捍卫冲突的模糊文本解释，暴露单智能体模型所遗漏的逻辑细微差别。在UKP学生论文语料库上的评估显示，MAD-ACC在无需领域特定训练的情况下，实现了85.7%的宏F1分数，显著优于单智能体推理基线。此外，与'黑盒'分类器不同，MAD-ACC的辩证方法通过生成可读的辩论记录，提供透明且可解释的替代方案。
英文摘要：
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.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
