Research map / Label-efficient and robust learning / Recognition and long-tail
Towards Personalized Review Summarization via User-Aware Sequence Network
Junjie Li, Haoran Li, Chengqing Zong
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 all four user-based strategies to S2S+Att (USN) | S2S+Att · Trip test | ROUGE-1: 22.09 -> 24.78 | Improved |
| Use USN instead of PGN | PGN · Trip test | ROUGE-1: 22.51 -> 24.78 | Improved |
| Add SEASS selective mechanism to S2S+Att | S2S+Att · Trip test | ROUGE-1: 22.09 -> 21.77 | Hurt |
| Add copy mechanism to S2S+Att (PGN) | S2S+Att · Trip test | ROUGE-1: 22.09 -> 22.51 | Improved |
| Use USN instead of PGN for aspect coverage | PGN · Trip human aspect eval | aspect F1: 0.530 -> 0.594 | Improved |
| Add User Selection strategy (ablation) | S2S+Att · Trip test | ROUGE-1: 21.15 -> 22.29 | Improved |
| Add User Prediction strategy (ablation) | S2S+Att · Trip test | ROUGE-1: 21.15 -> 24.04 | Improved |
| Add User Memory Prediction strategy (ablation) | S2S+Att · Trip test | ROUGE-1: 21.15 -> 23.36 | Improved |
| Add User Memory Generation strategy (ablation) | S2S+Att · Trip test | ROUGE-1: 21.15 -> 21.98 | Improved |
| Add User Selection to the other three strategies (ablation) | USN · Trip test | ROUGE-1: 23.23 -> 24.78 | Improved |
| Add User Prediction to the other three strategies (ablation) | USN · Trip test | ROUGE-1: 24.64 -> 24.78 | Improved |
| Add User Memory Prediction to the other three strategies (ablation) | USN · Trip test | ROUGE-1: 24.11 -> 24.78 | Improved |
Closest papers by meaning
- PersonalSum: A User-Subjective Guided Personalized Summarization Dataset for Large Language Models (NeurIPS 2024 · similarity 0.87)
- MeanSum: A Neural Model for Unsupervised Multi-Document Abstractive Summarization (ICML 2019 · similarity 0.86)
- Abstractive Text Summarization by Incorporating Reader Comments (AAAI 2019 · similarity 0.86)
- Tell Me What Is Good about This Property: Leveraging Reviews for Segment-Personalized Image Collection Summarization (AAAI 2024 · similarity 0.83)
- SummPilot: Bridging Efficiency and Customization for Interactive Summarization System (AAAI 2025 · similarity 0.82)
- Learning from Failure: Improving Meeting Summarization without Good Samples (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.
