说明: We describe the WikiQA dataset, a new publicly available set of question and sentence pairs, collected and annotated for research on open-domain question answering. Most previous work on answer sentence selection focuses on a dataset created using t <guotong1988> 上传 | 大小:6mb