Search Articles

View query in Help articles search

Search Results (1 to 10 of 13 Results)

Download search results: CSV END BibTex RIS


Exploiting Intersentence Information for Better Question-Driven Abstractive Summarization: Algorithm Development and Validation

Exploiting Intersentence Information for Better Question-Driven Abstractive Summarization: Algorithm Development and Validation

Wang et al [19] and Durmus et al [20] propose the QA-based factual consistency evaluation metrics QAGS and FEQA separately. They first generate a set of questions about the summary and then use a QA model to answer these questions for evaluation. Because of the characteristics of the Pub Med QA data set, the questions are general questions, and they can be answered by yes or no. We use the summaries as the context for the QA task to evaluate the factual consistency.

Xin Wang, Jian Wang, Bo Xu, Hongfei Lin, Bo Zhang, Zhihao Yang

JMIR Med Inform 2022;10(8):e38052