Text-to-Text Generation for Summarizing Informal Genres

该项目旨在通过使用文本到文本生成方法生成连贯和符合目标的摘要和答案，这是一种从输入文本生成新句子、融合相关短语并丢弃不相关短语的方法。正在开发一种用于文本到文本生成的句法统计框架，该框架可应用于非正式体裁，如转录语音和电子邮件，在这些体裁中，句子不能保证是完整的或符合语法的。正是这些流派将从这种方法中受益最大；对他们来说，单独使用句子摘要不是一种选择。目标是一个完全开发的句法统计框架，用于文本到文本的生成，其特点是在统计框架内使用完整的句法语法进行压缩和组合，一个将语用和语义的约束纳入生成系统的模型，能够从零碎和非语法的输入生成流利的、有语法意义的句子，以及从输入的文档句子生成高级抽象的句子的能力。该项目的特点是将压缩和语言模型集成到一个词汇化的中心驱动框架中，使生成器能够保持句子的语法，并避免显著改变意义的措辞变化。它的框架可以包含除语法之外的任意数量的特性，这些特性对于摘要很重要。一种新的动态编程技术允许从摘要/文档语料库中自动提取大量训练数据。关于谁与谁交谈和释义规则的信息将增加可以解决的修订范围。

This project aims at the generation of coherent and on-target summaries and answers through the use of text-to-text generation, an approach which generates new sentences from the input text, fusing relevant phrases and discarding irrelevant ones. A syntactic,statistical framework for text-to-text generation is being developed that can be applied to informal genres, such as transcribed speech and email, where sentences are not guaranteed to be either complete or grammatical. It is exactly these genres that stand to benefit the most from this approach; for them, summarization using sentence extractionalone is not an option.The aim is a fully developed, syntactic statistical framework for text-to-text generation which features the use of a full syntactic grammar within a statistical framework for compression and combination, a model for incorporating constraints from pragmatics andsemantics into the generation system, the ability to produce fluent, grammatical sentences from fragmentary and ungrammatical input, and the ability to generate sentences that make high level abstractions from input document sentences.The project features the integration of compression and language models into a lexicalized head-driven framework, enabling the generator to keep the sentence grammatical and avoid wording changes that dramatically alter meaning. Its framework can incorporate an arbitrary number of features beyond syntax that are important forsummarization. A new dynamic programming technique allows the automatic extraction of large amounts of training data from a summary/document corpus. Information about who speaks to whom and paraphrasing rules will increase the range of revisions that can be addressed.