Research map / Label-efficient and robust learning / Recognition and long-tail
Improving Multi-Document Summarization via Text Classification
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 text-classification data to learn document representations (ablation) | TCSum · DUC 2001 | ROUGE-1: 33.45 -> 36.45 | Helped |
| add text-classification data to learn document representations (ablation) | TCSum · DUC 2002 | ROUGE-1: 34.02 -> 36.90 | Helped |
| add text-classification data to learn document representations (ablation) | TCSum · DUC 2004 | ROUGE-1: 35.66 -> 38.27 | Helped |
| use category-specific transformation matrices instead of one shared matrix (ablation) | TCSum · DUC 2001 | ROUGE-1: 35.22 -> 36.45 | Helped |
| use category-specific transformation matrices instead of one shared matrix (ablation) | TCSum · DUC 2002 | ROUGE-1: 36.54 -> 36.90 | Helped |
| use category-specific transformation matrices instead of one shared matrix (ablation) | TCSum · DUC 2004 | ROUGE-1: 37.94 -> 38.27 | Helped |
| learn a summary embedding rather than rank by similarity to the document embedding (ablation) | TCSum · DUC 2001 | ROUGE-1: 24.66 -> 36.45 | Helped |
| learn a summary embedding rather than rank by similarity to the document embedding (ablation) | TCSum · DUC 2002 | ROUGE-1: 29.46 -> 36.90 | Helped |
| learn a summary embedding rather than rank by similarity to the document embedding (ablation) | TCSum · DUC 2004 | ROUGE-1: 30.80 -> 38.27 | Helped |
| use TCSum instead of R2N2 | DUC 2001 summarization · DUC 2001 | ROUGE-1: 35.88 -> 36.45 | Helped |
| use TCSum instead of the strongest listed baseline | DUC 2001 summarization · DUC 2001 | ROUGE-2: 7.86 -> 7.66 | Hurt |
| use TCSum instead of R2N2 | DUC 2002 summarization · DUC 2002 | ROUGE-1: 36.84 -> 36.90 | Helped |
Closest papers by meaning
- QuerySum: A Multi-Document Query-Focused Summarization Dataset Augmented with Similar Query Clusters (AAAI 2024 · similarity 0.84)
- FedSum: Data-Efficient Federated Learning Under Data Scarcity Scenario for Text Summarization (AAAI 2025 · similarity 0.83)
- Automatic Summarization of Long Documents (Student Abstract) (AAAI 2025 · similarity 0.83)
- Language Models of Code Are Few-Shot Planners and Reasoners for Multi-Document Summarization with Attribution (AAAI 2025 · similarity 0.82)
- MultiSum: A Multi-Facet Approach for Extractive Social Summarization Utilizing Semantic and Sociological Relationships (AAAI 2024 · similarity 0.82)
- Unsupervised Extractive Summarization with Learnable Length Control Strategies (AAAI 2024 · similarity 0.81)
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.