Exploring Discourse Structures for Argument Impact Classification
This work addresses argument impact classification for debate analysis, but it is incremental as it builds on existing tasks and methods.
The paper tackles the problem of classifying argument impact by showing that discourse relations between arguments are essential factors for identifying persuasive power, and proposes DisCOC to fuse discourse information with language model features, achieving improved performance on the task defined by Durmus et al. (2019).
Discourse relations among arguments reveal logical structures of a debate conversation. However, no prior work has explicitly studied how the sequence of discourse relations influence a claim's impact. This paper empirically shows that the discourse relations between two arguments along the context path are essential factors for identifying the persuasive power of an argument. We further propose DisCOC to inject and fuse the sentence-level structural discourse information with contextualized features derived from large-scale language models. Experimental results and extensive analysis show that the attention and gate mechanisms that explicitly model contexts and texts can indeed help the argument impact classification task defined by Durmus et al. (2019), and discourse structures among the context path of the claim to be classified can further boost the performance.