Abstract
Argument mining is a subfield of argumentation that aims to automatically extract argumentative structures and their relations from natural language texts. This paper investigates how a single large language model can be leveraged to perform one or several argument mining tasks. Our contributions are two-fold. First, we construct a multi-task dataset by surveying and converting 19 well-known argument mining datasets from the literature into a unified format. Second, we explore various training strategies using Meta AI's Llama-3.1-8B-Instruct model: (1) fine-tuning on individual tasks, (2) fine-tuning jointly on multiple tasks, and (3) merging models fine-tuned separately on individual tasks. Our experiments show that task-specific fine-tuning significantly improves individual performance across all tasks. Moreover, multi-task fine-tuning maintains strong performance without degradation, suggesting effective transfer learning across related tasks. Finally, we demonstrate that model merging offers a viable compromise: it yields competitive performance while mitigating the computational costs associated with full multi-task fine-tuning.
Abstract (translated)
论证挖掘是论辩学的一个子领域，旨在自动从自然语言文本中提取论证结构及其关系。本文探讨了如何利用单一的大规模语言模型来执行一项或多项论证挖掘任务。我们的贡献主要有两个方面：首先，我们通过调查并转换文献中的19个知名论证挖掘数据集，构建了一个多任务数据集，并将其统一为一种格式。其次，我们探索了使用Meta AI的Llama-3.1-8B-Instruct模型的各种训练策略：(1)在单一任务上进行微调；(2)在多个任务上联合进行微调；以及(3)将单独针对每个任务进行微调后的模型合并。我们的实验表明，特定任务的微调显著提升了所有任务上的个体性能。此外，多任务微调能够保持强劲的表现而不会出现退化现象，这表明跨相关任务的有效迁移学习。最后，我们证明了模型合并提供了一个可行的折中方案：它在降低完整多任务微调相关的计算成本的同时还能达到竞争性的表现。
URL
https://arxiv.org/abs/2508.17926
