<html><head><meta http-equiv="Content-Type" content="text/html; charset=gb2312"></head><body style="word-wrap: break-word; -webkit-nbsp-mode: space; line-break: after-white-space;" class=""><span style="font-size: 14px;" class="">Dear Adapters,<br class=""> In this week, I will give you the talk about the paper </span><font size="4" class="">ˇ°HIBERT: Document Level Pre-training of Hierarchical Bidirectional
Transformers for Document Summarizationˇ±</font><span style="font-size: 14px;" class=""> which is accepted by ACL in 2019. </span><div class=""><span style="font-size: 14px;" class="">In this paper,t</span><span style="font-family: NimbusRomNo9L; font-size: 11pt;" class="">he core part of a neural extractive summarization model is the hierarchical document encoder. And author proposed a method to pre-train document level
hierarchical bidirectional transformer encoders on
unlabeled data. The HIBERT model achieves a quite good performance in CNNDM dataset and NYT50 dataset. </span><span style="font-size: 14px;" class=""><br class=""></span><br class=""><br class=""><span style="font-size: 14px;" class="">Time: Wed 5:00pm</span><br class=""><span style="font-size: 14px;" class="">Venue: SEIEE 3-414</span><br class=""><br class=""><span style="font-size: 14px;" class="">Best regards, </span><br class=""><span style="font-size: 14px;" class="">Sophia</span></div></body></html>