QBSUM: A large-scale query-based document summarization dataset from real-world applications

Query-based document summarization aims to extract or generate a summary of which directly answers is relevant the search query. It an important technique that can be beneficial variety applications such as engines, document-level machine reading comprehension, and chatbots. Currently, datasets designed for query-based are short in numbers existing also limited both scale quality. Moreover, best our knowledge, there no publicly available dataset Chinese summarization. In this paper, we present QBSUM, high-quality large-scale consisting 49,000+ data samples task We propose multiple unsupervised supervised solutions demonstrate their high-speed inference superior performance via offline experiments online A/B tests. The QBSUM released order facilitate future advancement research field.