Evaluating LLM-Driven Summarisation of Parliamentary Debates with Computational Argumentation

ArXi:2604.19331v1 Announce Type: new Understanding how policy is debated and justified in parliament is a fundamental aspect of the cratic process. However, the volume and complexity of such debates mean that outside audiences struggle to engage. Meanwhile, Large Language Models (LLMs) have been shown to enable automated summarisation at scale. While summaries of debates can make parliamentary procedures accessible, evaluating whether these summaries faithfully communicate argumentative content remains challenging.