arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1904.07142cs.CL
收藏

基于半马尔可夫条件反射函数的神经网络章节标题生成方法提高人类文本理解能力

Improving Human Text Comprehension through Semi-Markov CRF-based Neural Section Title Generation

Harvard SEAS(哈佛大学工程与应用科学学院);Adobe Research(奥多比研究院);University of Illinois Urbana Champaign(伊利诺伊大学厄巴纳-香槟分校)

Sebastian Gehrmann, Steven Layne, Franck Dernoncourt

摘要:

长文档中的短部分标题通过引导读者将注意力集中在相关段落并提供有助于理解文档进展的锚点来支持读者。当对阅读能力较低的读者进行测量时,章节标题的积极影响甚至更加明显,例如在标签文本资源有限的社区中。因此,我们的目标是开发技术,在低资源环境中生成章节标题。特别是,我们提出了一个提取管道的章节标题生成,首先选择最突出的句子,然后应用基于删除的压缩。我们的压缩方法是基于半马尔可夫条件随机场,利用无监督的字表示,如ELMo或BERT,消除了对复杂的编码器-解码器架构的需要。结果表明,这种方法导致竞争力的性能与序列到序列模型与高资源,而强烈优于它与低资源。在一项针对不同阅读能力的受试者的人类受试者研究中,我们发现,我们的章节标题提高了完成理解任务的速度,同时保持了类似的准确性。

英文摘要:

Titles of short sections within long documents support readers by guiding their focus towards relevant passages and by providing anchor-points that help to understand the progression of the document. The positive effects of section titles are even more pronounced when measured on readers with less developed reading abilities, for example in communities with limited labeled text resources. We, therefore, aim to develop techniques to generate section titles in low-resource environments. In particular, we present an extractive pipeline for section title generation by first selecting the most salient sentence and then applying deletion-based compression. Our compression approach is based on a Semi-Markov Conditional Random Field that leverages unsupervised word-representations such as ELMo or BERT, eliminating the need for a complex encoder-decoder architecture. The results show that this approach leads to competitive performance with sequence-to-sequence models with high resources, while strongly outperforming it with low resources. In a human-subject study across subjects with varying reading abilities, we find that our section titles improve the speed of completing comprehension tasks while retaining similar accuracy.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑