Research map / Label-efficient and robust learning / Recognition and long-tail
Persuasive Influence Detection: The Role of Argument Sequencing
Christopher Hidey, Kathleen McKeown
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 frame and discourse embeddings to word-LSTM | word-LSTM · CMV influence prediction, full path | AUC: 75.1 -> 75.5 | Improved |
| add highway network (ablation) | all-LSTM · CMV influence prediction, full path | accuracy: 70.1 -> 75.1 | Improved |
| add bi-LSTM sequence modeling (ablation) | all-LSTM · CMV influence prediction, full path | AUC: 73.2 -> 75.5 | Improved |
| add attention over LSTM states (ablation) | all-LSTM · CMV influence prediction, full path | AUC: 74.5 -> 75.5 | Improved |
| add original-post context via memory network | all-LSTM · CMV influence prediction, root reply | accuracy: 72.5 -> 75.0 | Improved |
| add original-post context via memory network | all-LSTM · CMV influence prediction, full path | accuracy: 75.1 -> 74.3 | Mixed |
| add interplay features to LSTM+memory | all-LSTM+memory · CMV influence prediction, full path | accuracy: 74.3 -> 81.0 | Improved |
| add original-post context in pairwise task | all-LSTM · CMV pairwise, full path | accuracy: 71.5 -> 71.6 | No clear effect |
| use only discourse embeddings (ablation) | all-LSTM · CMV influence prediction, full path | AUC: 75.5 -> 63.6 | Hurt |
| use only frame embeddings (ablation) | all-LSTM · CMV influence prediction, full path | AUC: 75.5 -> 66.4 | Hurt |
| LSTM model vs BoW | all-LSTM · CMV malleability prediction | accuracy: 51.6 -> 58.4 | Improved |
| model vs novice human annotators | all-LSTM+IP · CMV pairwise | accuracy: 54.84 -> 71.99 | Improved |
Closest papers by meaning
- Weakly-Supervised Hierarchical Models for Predicting Persuasive Strategies in Good-faith Textual Requests (AAAI 2021 · similarity 0.89)
- Measuring And Improving Persuasiveness Of Large Language Models (ICLR 2025 · similarity 0.87)
- Debating with More Persuasive LLMs Leads to More Truthful Answers (ICML 2024 · similarity 0.84)
- Beyond Detection: Exploring Evidence-based Multi-Agent Debate for Misinformation Intervention and Persuasion (AAAI 2026 · similarity 0.83)
- OpenDebateEvidence: A Massive-Scale Argument Mining and Summarization Dataset (NeurIPS 2024 · similarity 0.82)
- A Persuasive Approach to Combating Misinformation (ICML 2024 · similarity 0.82)
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.
