Summarization Programs: Interpretable Abstractive Summarization with Neural Modular Trees-猫眼课题宝

DOI：

10.48550/arxiv.2209.10492

发表时间：

2022-09

期刊：

ArXiv

影响因子：

通讯作者：

中科院分区：

其他

文献类型：

作者：

Swarnadeep Saha;Shiyue Zhang;Peter Hase;Mohit Bansal

当前的抽象摘要模型要么缺乏清晰的可解释性，要么只突出显示源文档的一部分，从而提供不完整的基本原理。为此，我们提出了总结程序（SP），一个可解释的模块化框架，由一个（有序）的二叉树列表组成，每个编码的一步一步的生成过程中的抽象摘要句子从源文件。摘要程序包含每个摘要句子的一个根节点，并且不同的树将每个摘要句子（根节点）连接到它所源自的文档句子（叶节点），连接节点包含中间生成的句子。边表示在摘要中涉及的不同模块化操作，例如句子融合、压缩和释义。我们首先提出了一个有效的最佳优先搜索方法，在神经模块，SP-Search，识别SP的人类摘要直接优化ROUGE分数。接下来，使用这些程序作为自动监督，我们提出了seq 2seq模型，生成摘要程序，然后执行这些程序以获得最终摘要。我们证明，SP-Search有效地代表了人类总结背后的生成过程，使用通常忠实于其预期行为的模块。我们还进行了模拟研究，以表明摘要程序提高了摘要模型的可解释性，允许人类更好地模拟模型推理。摘要程序是迈向可解释和模块化抽象摘要的有希望的一步，这是一项复杂的任务，以前主要通过黑盒端到端神经系统来解决。支持代码可在https://github.com/swarnaHub/SummarizationPrograms上获得

Current abstractive summarization models either suffer from a lack of clear interpretability or provide incomplete rationales by only highlighting parts of the source document. To this end, we propose the Summarization Program (SP), an interpretable modular framework consisting of an (ordered) list of binary trees, each encoding the step-by-step generative process of an abstractive summary sentence from the source document. A Summarization Program contains one root node per summary sentence, and a distinct tree connects each summary sentence (root node) to the document sentences (leaf nodes) from which it is derived, with the connecting nodes containing intermediate generated sentences. Edges represent different modular operations involved in summarization such as sentence fusion, compression, and paraphrasing. We first propose an efficient best-first search method over neural modules, SP-Search that identifies SPs for human summaries by directly optimizing for ROUGE scores. Next, using these programs as automatic supervision, we propose seq2seq models that generate Summarization Programs, which are then executed to obtain final summaries. We demonstrate that SP-Search effectively represents the generative process behind human summaries using modules that are typically faithful to their intended behavior. We also conduct a simulation study to show that Summarization Programs improve the interpretability of summarization models by allowing humans to better simulate model reasoning. Summarization Programs constitute a promising step toward interpretable and modular abstractive summarization, a complex task previously addressed primarily through blackbox end-to-end neural systems. Supporting code available at https://github.com/swarnaHub/SummarizationPrograms

Summarization Programs: Interpretable Abstractive Summarization with Neural Modular Trees

文献摘要

相似文献