Research map / Large language models / Language, safety and interpretability
Argument Mining from Speech: Detecting Claims in Political Debates
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add MFCC audio features to text features (ablation) | SVM on ground-truth transcript · UK 2015 debate, Cameron | F1: 55.2 -> 61.2 | Helped |
| add MFCC audio features to text features (ablation) | SVM on ground-truth transcript · UK 2015 debate, Clegg | F1: 50.6 -> 59.0 | Helped |
| add MFCC audio features to text features (ablation) | SVM on ground-truth transcript · UK 2015 debate, Miliband | F1: 58.7 -> 62.5 | Helped |
| add MFCC audio features to text features (ablation) | SVM on Google Speech transcript · UK 2015 debate, Cameron | F1: 48.2 -> 52.6 | Helped |
| add MFCC audio features to text features (ablation) | SVM on Google Speech transcript · UK 2015 debate, Clegg | F1: 50.5 -> 52.5 | Helped |
| add MFCC audio features to text features (ablation) | SVM on Google Speech transcript · UK 2015 debate, Miliband | F1: 31.3 -> 29.3 | Hurt |
| use ASR (Google Speech) transcript instead of ground-truth transcript (ablation) | SVM claim detection · UK 2015 debate, Miliband | F1: 58.7 -> 31.3 | Hurt |
Closest papers by meaning
- Never Retreat, Never Retract: Argumentation Analysis for Political Speeches (AAAI 2018 · similarity 0.89)
- DISPUTool 2.0: A Modular Architecture for Multi-Layer Argumentative Analysis of Political Debates (AAAI 2023 · similarity 0.88)
- OpenDebateEvidence: A Massive-Scale Argument Mining and Summarization Dataset (NeurIPS 2024 · similarity 0.86)
- Debate or Vote: Which Yields Better Decisions in Multi-Agent Large Language Models? (NeurIPS 2025 · similarity 0.81)
- Lawma: The Power of Specialization for Legal Annotation (ICLR 2025 · similarity 0.81)
- The Power of LLM-Generated Synthetic Data for Stance Detection in Online Political Discussions (ICLR 2025 · similarity 0.80)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.