Research map / Vision-language and multimodal / Visual question answering
Towards Improving Faithfulness in Abstractive 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 |
|---|---|---|---|
| Add the FES QA multi-task framework and QA-guided decoder (ablation) | BART summarization on CNN/DM · CNN/DM | ROUGE-L: 41.61 -> 43.47 | Helped |
| Add QA attention guidance to the decoder (ablation) | FES on CNN/DM · CNN/DM | QuestEval: 35.22 -> 35.50 | Helped |
| Add the max-margin loss (ablation) | FES on CNN/DM · CNN/DM | FactCC: 54.60 -> 55.23 | Helped |
| Select QA pairs about key information rather than random QA pairs (ablation) | FES on CNN/DM · CNN/DM | QuestEval: 34.78 -> 35.50 | Helped |
| Increase the number of oracle QA pairs from fewer than 8 to 8 (ablation) | FES on CNN/DM · CNN/DM | ROUGE | Helped |
| Increase the number of QA pairs beyond 8 (ablation) | FES on CNN/DM · CNN/DM | ROUGE | No clear effect |
| Use FES instead of the strongest baseline | Summarization systems · CNN/DM | FactCC: 53.51 -> 55.23 | Helped |
| Use FES instead of the strongest baseline | Summarization systems · XSum | QuestEval: 22.97 -> 25.83 | Helped |
Closest papers by meaning
- A Deep Reinforced Model for Abstractive Summarization (ICLR 2018 · similarity 0.90)
- Faithful to the Original: Fact Aware Neural Abstractive Summarization (AAAI 2018 · similarity 0.90)
- Learning to Summarize by Learning to Quiz: Adversarial Agentic Collaboration for Long Document Summarization (ICLR 2026 · similarity 0.88)
- Distilling Structured Rationale from Large Language Models to Small Language Models for Abstractive Summarization (AAAI 2025 · similarity 0.86)
- Improving Faithfulness in Abstractive Text Summarization with EDUs Using BART (Student Abstract) (AAAI 2024 · similarity 0.86)
- Learning from Failure: Improving Meeting Summarization without Good Samples (AAAI 2024 · similarity 0.86)
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.