A Survey on Cross-Lingual Summarization

Abstract Cross-lingual summarization is the task of generating a summary in one language (e.g., English) for given document(s) different Chinese). Under globalization background, this has attracted increasing attention computational linguistics community. Nevertheless, there still remains lack comprehensive review task. Therefore, we present first systematic critical on datasets, approaches, and challenges field. Specifically, carefully organize existing datasets approaches according to construction methods solution paradigms, respectively. For each type dataset or approach, thoroughly introduce summarize previous efforts further compare them with other provide deeper analyses. In end, also discuss promising directions offer our thoughts facilitate future research. This survey both beginners experts cross-lingual summarization, hope it will serve as starting point well source new ideas researchers engineers interested area.