提出新解码算法，精准控制摘要长度并生成更优结果。
A Decoding Algorithm for Length-Control Summarization Based on Directed Acyclic Transformers
- 基于有向无环变换器，通过路径预测实现长度可控
- 在Gigaword和DUC2004上达到当前最优性能
- 适合需要精确长度控制的摘要生成场景
长度控制摘要旨在将长文本压缩至指定长度范围内。以往方法多采用自回归模型，将长度要求视为软约束，难以保证严格满足。本文提出一种基于有向无环变换器（DAT）的新解码算法，支持多个合理文本片段并预测连接它们的路径。同时，提出序列最大后验（SeqMAP）解码算法，对不同路径进行边缘化，寻找最可能且符合长度预算的摘要。该算法基于束搜索，进一步支持重排序以提升性能。在Gigaword与DUC2004数据集上的实验表明，该方法在长度控制摘要任务中达到当前最优水平。
原文摘要 · Abstract (English)
Length-control summarization aims to condense long texts into a short one within a certain length limit. Previous approaches often use autoregressive (AR) models and treat the length requirement as a soft constraint, which may not always be satisfied. In this study, we propose a novel length-control decoding algorithm based on the Directed Acyclic Transformer (DAT). Our approach allows for multiple plausible sequence fragments and predicts a \emph{path} to connect them. In addition, we propose a Sequence Maximum a Posteriori (SeqMAP) decoding algorithm that marginalizes different possible paths and finds the most probable summary satisfying the length budget. Our algorithm is based on beam search, which further facilitates a reranker for performance improvement. Experimental results on the Gigaword and DUC2004 datasets demonstrate our state-of-the-art performance for length-control summarization.
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。