论文标题
限制的抽象摘要：保持事实的一致性与受约束的生成
Constrained Abstractive Summarization: Preserving Factual Consistency with Constrained Generation
论文作者
论文摘要
尽管取得了重大进展，但最先进的抽象摘要方法仍然容易幻觉与源文档不一致。在本文中，我们提出了受约束的抽象摘要（CAS），这是一种一般设置，通过将令牌指定为摘要中必须存在的约束来保留抽象性摘要的事实一致性。我们采用词汇约束解码，这是一种通常适用于自回归生成模型的技术，以在两种情况下实现CAS并进行实验：（1）自动摘要而无需人类参与，其中从源文档中提取了键形，并用作约束； （2）人类引导的互动摘要，其中使用手动约束形式的人类反馈来指导摘要产生。在两个基准数据集上的自动和人类评估表明，CAS可以改善词汇叠加（Rouge）和抽象摘要的事实一致性。特别是，当在交互式汇总中仅使用一个手动约束时，我们最多可观察到13.8 Rouge-2。
Despite significant progress, state-of-the-art abstractive summarization methods are still prone to hallucinate content inconsistent with the source document. In this paper, we propose Constrained Abstractive Summarization (CAS), a general setup that preserves the factual consistency of abstractive summarization by specifying tokens as constraints that must be present in the summary. We adopt lexically constrained decoding, a technique generally applicable to autoregressive generative models, to fulfill CAS and conduct experiments in two scenarios: (1) automatic summarization without human involvement, where keyphrases are extracted from the source document and used as constraints; (2) human-guided interactive summarization, where human feedback in the form of manual constraints are used to guide summary generation. Automatic and human evaluations on two benchmark datasets demonstrate that CAS improves both lexical overlap (ROUGE) and factual consistency of abstractive summarization. In particular, we observe up to 13.8 ROUGE-2 gains when only one manual constraint is used in interactive summarization.
