arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1910.12840cs.CL
收藏

文摘文本摘要的事实一致性评价

Evaluating the Factual Consistency of Abstractive Text Summarization

Salesforce Research(赛富时研究院)

Wojciech Kryściński, Bryan McCann, Caiming Xiong, Richard Socher

摘要:

目前用于评估摘要算法的指标并不考虑摘要是否与源文档事实上一致。我们提出了一个弱监督,基于模型的方法来验证事实的一致性和识别源文件和生成的摘要之间的冲突。训练数据是通过对源文档的句子应用一系列基于规则的转换来生成的。事实一致性模型然后被联合训练用于三个任务:1)识别句子在变换后是否保持事实一致,2)在源文档中提取跨度以支持一致性预测,3)在摘要句子中提取不一致的跨度(如果存在的话)。将此模型转换为由几个最先进的模型生成的摘要表明,这种高度可扩展的方法大大优于以前的模型,包括使用标准数据集进行自然语言推理和事实检查的强监督训练的模型。此外,人类的评价表明,辅助跨度提取任务提供了有用的帮助,在验证事实的一致性的过程中。

英文摘要:

Currently used metrics for assessing summarization algorithms do not account for whether summaries are factually consistent with source documents. We propose a weakly-supervised, model-based approach for verifying factual consistency and identifying conflicts between source documents and a generated summary. Training data is generated by applying a series of rule-based transformations to the sentences of source documents. The factual consistency model is then trained jointly for three tasks: 1) identify whether sentences remain factually consistent after transformation, 2) extract a span in the source documents to support the consistency prediction, 3) extract a span in the summary sentence that is inconsistent if one exists. Transferring this model to summaries generated by several state-of-the art models reveals that this highly scalable approach substantially outperforms previous models, including those trained with strong supervision using standard datasets for natural language inference and fact checking. Additionally, human evaluation shows that the auxiliary span extraction tasks provide useful assistance in the process of verifying factual consistency.

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。

↑