Towards Content Transfer through Grounded Text Generation
Shrimai Prabhumoye Carnegie Mellon University 5000 Forbes Avenue Pittsburgh, PA 15219、Chris Quirk, Michel Galley Microsoft Research One Microsoft Way Redmond, WA 98052
Abstract
Recent work in neural generation has attracted significant interest in controlling the form of text, such as style, persona, and politeness. However, there has been less work on controlling neural text generation for content. This paper introduces the notion of Content Transfer for long-form text generation, where the task is to generate a next sentence in a document that both fits its context and is grounded in a content-rich external textual source such as a news story. Our experiments on Wikipedia data show significant improvements against competitive baselines. As another contribution of this paper, we release a benchmark dataset of 640k Wikipedia referenced sentences paired with the source articles to encourage exploration of this new task.
中文速览
长篇文本生成中一个棘手难题是：如何让模型生成的句子既契合上下文语境，又能忠实地反映外部信息源（如新闻报道）的真实内容，而不是凭空捏造。这篇论文提出了"内容迁移"（Content Transfer）这一新任务框架，要求模型以现有文档片段为上下文、以外部新闻文章为内容依据，自动生成下一句连贯且有实质内容的更新句。研究者构建了一个包含64万条样本的维基百科基准数据集——每条样本对应一句维基百科引用句及其所引用的新闻原文——并设计了多种生成式和抽取式模型，其中同时编码上下文与外部文档的模型（CRG）比单纯做摘要的基线方法高出7个ROUGE-L分。这项工作的价值在于：它为写作辅助工具提供了一条可靠路径，让AI在帮助作者续写时既不说废话、也不胡编乱造，而是有据可依地融入外部知识。
原文 arXiv:1905.05293;中英对照 + 大白话阅读 https://aha.fim.ai/paper/1905.05293v1