Research map / Architectures and efficiency / Attention and transformers
A Character-Level Length-Control Algorithm for Non-Autoregressive Sentence Summarization
Puyuan Liu, Xiang Zhang, Lili Mou
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 |
|---|---|---|---|
| replace truncation with character-level DP length control (ablation) | NACC · Gigaword test, supervised | ROUGE-1 F1: 33.12 -> 33.66 | Improved |
| replace truncation with character-level DP length control (ablation) | NACC · Gigaword test, unsupervised | ROUGE-1 F1: 25.79 -> 27.45 | Improved |
| replace truncation with character-level DP length control (ablation) | NACC · DUC2004 | ROUGE-1 Recall: 26.43 -> 28.37 | Improved |
| replace truncation with character-level DP length control (ablation) | Autoregressive Transformer · Gigaword test, supervised | ROUGE-1 F1: 35.12 -> 34.10 | Hurt |
| replace truncation with character-level DP length control (ablation) | Autoregressive Transformer · Gigaword test, unsupervised | ROUGE-1 F1: 26.31 -> 25.33 | Hurt |
| use character-constrained search instead of word-constrained search followed by truncation | Schumann et al. search · Gigaword test, unsupervised | ROUGE-1 F1: 24.98 -> 25.30 | Improved |
| increase the length bucket size (ablation) | NACC length-control decoding · Gigaword test, unsupervised | mean ROUGE | Hurt |
| use character-length control rather than truncation (ablation) | NACC · Gigaword test, unsupervised human evaluation | overall quality win rate: 18% -> 38% | Improved |
| use character-length control rather than truncation (ablation) | NACC · Gigaword test, unsupervised human evaluation | completeness and fluency win rate: 22% -> 42% | Improved |
Closest papers by meaning
- Unsupervised Extractive Summarization with Learnable Length Control Strategies (AAAI 2024 · similarity 0.86)
- Flexible Non-Autoregressive Extractive Summarization with Threshold: How to Extract a Non-Fixed Number of Summary Sentences (AAAI 2021 · similarity 0.86)
- Joint Parsing and Generation for Abstractive Summarization (AAAI 2020 · similarity 0.85)
- Automatic Summarization of Long Documents (Student Abstract) (AAAI 2025 · similarity 0.85)
- Learning to Summarize by Learning to Quiz: Adversarial Agentic Collaboration for Long Document Summarization (ICLR 2026 · similarity 0.84)
- Distilling Structured Rationale from Large Language Models to Small Language Models for Abstractive Summarization (AAAI 2025 · similarity 0.83)
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.
