New Alignment Methods for Discriminative Book Summarization Work in Progress
David Bamman Noah A. Smith School of Computer Science Carnegie Mellon University Pittsburgh, PA 15213, USA
Abstract
We consider the unsupervised alignment of the full text of a book with a human-written summary. This presents challenges not seen in other text alignment problems, including a disparity in length and, consequent to this, a violation of the expectation that individual words and phrases should align, since large passages and chapters can be distilled into a single summary phrase. We present two new methods, based on hidden Markov models, specifically targeted to this problem, and demonstrate gains on an extractive book summarization task. While there is still much room for improvement, unsupervised alignment holds intrinsic value in offering insight into what features of a book are deemed worthy of summarization.
中文速览
把一整本书与人写的情节摘要自动对齐，难点在于两者篇幅悬殊——书平均是摘要的80多倍长，传统的词/短语级对齐根本不够用，因为书里整章的内容可能只对应摘要里的一句话。作者提出了两种基于隐马尔可夫模型（Hidden Markov Model, HMM）的无监督对齐方法：一种是"段落模型"，让每个隐状态对应原书中长度可变的一段连续文字，用该段的一元语言模型概率来衡量它与摘要句的匹配程度，并通过采样来动态调整段落边界；另一种是"词元模型"，在经典HMM词对齐框架上引入参数学习和空对齐机制，以应对长文档中稀疏对齐的挑战。在基于Project Gutenberg书籍与Wikipedia情节摘要构建的数据集上，两种方法在抽取式摘要任务中均优于基线，证明了无监督对齐在书籍摘要场景下的可行性，也为理解"书里哪些内容值得被摘要"提供了有价值的分析工具。
原文 arXiv:1305.1319;中英对照 + 大白话阅读 https://aha.fim.ai/paper/1305.1319v1