将生成与筛选分离，提升摘要事实一致性。
Decoupling Generation and Selection for Budget-Constrained Faithful Summarization
- 先生成多条候选摘要，再通过组合优化筛选最佳结果。
- 在多文档摘要上显著提升事实准确性和来源依据度。
- 适合需要高可信度摘要的场景，如新闻、科研报告。
抽象摘要模型仍易出现事实错误、重复和长度控制不佳问题。本文提出一种模块化生成-筛选框架，用于句子预算约束下的摘要生成。预训练生成器产出多个候选摘要，并分解为句子级候选；组合选择器在明确预算下，通过平衡相关性、事实性和冗余度，构建最终摘要。该框架支持MMR、ILP及受DPP启发的对数行列式目标，无需重新训练生成器。在CNN/DailyMail、Multi-News、FaithBench和TofuEval上的实验表明，事实性和来源依据度持续提升，尤其在多文档摘要任务中表现突出，代价是参考重叠度略低。人工评估显示摘要感知一致性、相关性、清晰度和简洁性更高，仅轻微降低连贯性。结果表明，解耦生成与选择是一种无需依赖特定模型的事实对齐机制。代码已公开于https://anonymous.4open.science/r/bcfs-D05E/。
原文摘要 · Abstract (English)
Abstractive summarization models remain vulnerable to factual inconsistency, redundancy, and weak length control. We propose a modular generation-and-selection framework for sentence-budget-constrained summarization. A pretrained generator produces multiple candidate summaries, which are decomposed into sentence-level candidates. A combinatorial selector then constructs the final summary by balancing relevance, factuality, and redundancy under an explicit budget. The framework supports MMR, ILP, and a DPP-inspired log-determinant objective without retraining the generator. Experiments on CNN/DailyMail, Multi-News, FaithBench, and TofuEval show consistent improvements in factuality and source-grounding metrics, especially for multi-document summarization, at the cost of lower reference-overlap scores. Human evaluation further indicates higher perceived consistency, relevance, clarity, and conciseness, with a small reduction in coherence. These results show that decoupling generation from selection provides a model-agnostic mechanism for improving factual grounding. Code is available at https://anonymous.4open.science/r/bcfs-D05E/.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。