Research map / Graph learning / Knowledge and temporal graphs
Hierarchical Heterogeneous Graph Attention Network for Syntax-Aware Summarization
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 |
|---|---|---|---|
| use constituency-tree syntax-aware hierarchical graph attention | SynapSum-Abs · CNN/DM | ROUGE-1 | Helped |
| use constituency-tree syntax-aware hierarchical graph attention | SynapSum-Ext · WikiHow | ROUGE-1 | Helped |
| use constituency-tree syntax-aware hierarchical graph attention | SynapSum-Abs · NYT | ROUGE-1 | Hurt |
| add pre-trained BART initialization and fine-tuning | SynapSum-Abs · CNN/DM | ROUGE-1: 41.52 -> 44.32 | Helped |
| add pre-trained MatchSum initialization and fine-tuning | SynapSum-Ext · NYT | ROUGE-1: 48.93 -> 50.30 | Helped |
| use a higher-quality constituency parser | SynapSum · CNN/DM | BERTScore | Helped |
| add bidirectional LSTM word-node embeddings instead of random initialization (ablation) | SynapSum-Abs · CNN/DM | ROUGE-1: 41.40 -> 41.52 | Helped |
| add the syntax-aware graph attention layer (ablation) | SynapSum-Abs · CNN/DM | ROUGE-1: 41.04 -> 41.52 | Helped |
| add the hierarchical graph pooling layer (ablation) | SynapSum-Abs · CNN/DM | ROUGE-1: 41.12 -> 41.52 | Helped |
| add the copying mechanism (ablation) | SynapSum-Abs · CNN/DM | ROUGE-1: 41.26 -> 41.52 | Helped |
Closest papers by meaning
- Structured Neural Summarization (ICLR 2019 · similarity 0.89)
- Summarization Programs: Interpretable Abstractive Summarization with Neural Modular Trees (ICLR 2023 · similarity 0.88)
- Unity in Diversity: Learning Distributed Heterogeneous Sentence Representation for Extractive Summarization (AAAI 2018 · similarity 0.88)
- Text summarization via global structure awareness (ICLR 2026 · similarity 0.87)
- Joint Parsing and Generation for Abstractive Summarization (AAAI 2020 · similarity 0.87)
- Language Models of Code Are Few-Shot Planners and Reasoners for Multi-Document Summarization with Attribution (AAAI 2025 · similarity 0.84)
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.