Research map / Label-efficient and robust learning / Recognition and long-tail
TGSum: Build Tweet Guided Multi-Document Summarization Dataset
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 TGSum training data to DUC training data | SVR summarizer · DUC 01 | ROUGE-1: 29.78 -> 30.32 | Helped |
| Add TGSum training data to DUC training data | SVR summarizer · DUC 01 | ROUGE-2: 6.01 -> 6.26 | Helped |
| Add TGSum training data to DUC training data | SVR summarizer · DUC 02 | ROUGE-1: 31.56 -> 32.15 | Helped |
| Add TGSum training data to DUC training data | SVR summarizer · DUC 02 | ROUGE-2: 6.78 -> 6.89 | Helped |
| Add TGSum training data to DUC training data | SVR summarizer · DUC 04 | ROUGE-1: 36.18 -> 36.64 | Helped |
| Add TGSum training data to DUC training data | SVR summarizer · DUC 04 | ROUGE-2: 9.34 -> 9.51 | Helped |
| Use ILP sentence selection instead of greedy selection | ROUGE-optimized summary generation · Linked-tweet ROUGE evaluation | ROUGE-1: 48.29 -> 55.24 | Helped |
| Use ILP sentence selection instead of greedy selection | ROUGE-optimized summary generation · Linked-tweet ROUGE evaluation | ROUGE-2: 29.44 -> 34.74 | Helped |
| Jointly optimize ROUGE-1 and ROUGE-2 rather than only ROUGE-2 (ablation) | ILP summary generation · Linked-tweet ROUGE evaluation | ROUGE-1: 49.15 -> 50.64 | Helped |
| Jointly optimize ROUGE-1 and ROUGE-2 rather than only ROUGE-2 (ablation) | ILP summary generation · Linked-tweet ROUGE evaluation | ROUGE-2: 34.74 -> 34.74 | No clear effect |
| Use TGSum references instead of raw linked tweets as extra training data | SVR summarizer · DUC 04 | ROUGE-2: 9.23 -> 9.51 | Helped |
Closest papers by meaning
- MultiSum: A Multi-Facet Approach for Extractive Social Summarization Utilizing Semantic and Sociological Relationships (AAAI 2024 · similarity 0.85)
- PersonalSum: A User-Subjective Guided Personalized Summarization Dataset for Large Language Models (NeurIPS 2024 · similarity 0.83)
- BooookScore: A systematic exploration of book-length summarization in the era of LLMs (ICLR 2024 · similarity 0.83)
- QuerySum: A Multi-Document Query-Focused Summarization Dataset Augmented with Similar Query Clusters (AAAI 2024 · similarity 0.83)
- Language Models of Code Are Few-Shot Planners and Reasoners for Multi-Document Summarization with Attribution (AAAI 2025 · similarity 0.82)
- Unsupervised Extractive Summarization with Learnable Length Control Strategies (AAAI 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.