发表机构
College of Applied Science, Shenzhen University; School of Artificial Intelligence, Shenzhen Technology University; Harbin Institute of Technology (Shenzhen); The Chinese University of Hong Kong; Pengcheng Laboratory（深圳大学应用科学学院; 深圳技术大学人工智能学院; 哈尔滨工业大学（深圳）; 香港中文大学; 鹏城实验室）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对论辩挖掘中高质量标注数据稀缺的问题，提出对抗强化学习框架，联合优化生成器与判别器，在保持多样性的同时提升合成数据的结构准确性，并在三个基准数据集上验证了有效性。
AI 中文摘要
论辩挖掘（Argument Mining, AM）从根本上受到高质量结构标注数据集稀缺的制约。尽管大语言模型（LLMs）在合成数据生成方面展现出潜力，但生成既在结构上准确又具有足够多样性的合成AM数据仍然是一个具有挑战性的问题。为解决这一问题，我们从新的视角重新审视AM的合成数据生成，并提出了一种新颖的对抗强化学习数据合成框架。该框架在对抗循环中联合优化生成器和判别器，其中生成器产生结构化的AM实例，判别器通过区分真实数据与合成候选来提供学习信号。这使得生成器能够通过对抗反馈逐步提高生成论辩数据的结构准确性，同时保持多样性。大量实验表明，所提出的框架在三个基准数据集上的全数据及低资源设置下均持续提升了AM性能，验证了其有效性和可扩展性。
英文摘要
Argument Mining (AM) is fundamentally constrained by the scarcity of high-quality structure-annotated datasets. While LLMs have shown promise in synthetic data generation, producing synthetic AM data that is both structurally accurate and sufficiently diverse remains a challenging problem. To address this problem, we revisit synthetic data generation for AM from a new perspective and propose a novel adversarial reinforcement learning framework for data synthesis. The proposed framework jointly optimizes the generator and the discriminator in an adversarial loop, in which the generator produces structured AM instances, and the discriminator provides learning signals by distinguishing real data from synthetic candidates. This enables the generator to progressively improve both the structural accuracy of generated argument data while maintaining diversity through adversarial feedback. Extensive experiments demonstrate that the proposed framework consistently improves AM performance on three benchmark datasets in both full-data and low-resource settings, validating its effectiveness and scalability.
CommentsAccepted to Findings of EMNLP 2026
