Research map / Probabilistic and scientific modeling / Variational and generative sampling
Salience Estimation via Variational Auto-Encoders for Multi-Document Summarization
Piji Li, Zihao Wang, Wai Lam, Zhaochun Ren, Lidong Bing
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 |
|---|---|---|---|
| use full VAEs-A (aspect reconstruction with alignment) instead of VAEs-Zero (ablation) | VAEs-based salience · DUC 2006 | ROUGE-2: 0.080 -> 0.089 | Improved |
| use full VAEs-A (aspect reconstruction with alignment) instead of VAEs-Zero (ablation) | VAEs-based salience · DUC 2007 | ROUGE-2: 0.106 -> 0.110 | Improved |
| use full VAEs-A (aspect reconstruction with alignment) instead of VAEs-Zero (ablation) | VAEs-based salience · TAC 2011 | ROUGE-2: 0.113 -> 0.122 | Improved |
| use VAEs-A salience instead of term-frequency salience (ABS-Phrase) in the phrase-based ILP summarizer | ABS-Phrase · DUC 2006 | ROUGE-2: 0.082 -> 0.089 | Improved |
| use VAEs-A salience instead of RA-MDS sparse coding | RA-MDS · TAC 2011 | ROUGE-2: 0.117 -> 0.122 | Improved |
| use simple VAEs-Zero salience instead of ABS-Phrase term frequency | ABS-Phrase · TAC 2011 | ROUGE-1: 0.393 -> 0.388 | Mixed |
Closest papers by meaning
- DeepChannel: Salience Estimation by Contrastive Learning for Extractive Document Summarization (AAAI 2019 · similarity 0.88)
- Document Summarization with VHTM: Variational Hierarchical Topic-Aware Mechanism (AAAI 2020 · similarity 0.85)
- MeanSum: A Neural Model for Unsupervised Multi-Document Abstractive Summarization (ICML 2019 · similarity 0.85)
- A Deep Reinforced Model for Abstractive Summarization (ICLR 2018 · similarity 0.85)
- Unity in Diversity: Learning Distributed Heterogeneous Sentence Representation for Extractive Summarization (AAAI 2018 · similarity 0.85)
- Learning to Summarize by Learning to Quiz: Adversarial Agentic Collaboration for Long Document Summarization (ICLR 2026 · 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.
