主题论文总结4：长文本摘要（持续更新ing...）_globalizing bert-based transformer architectures f

诸神缄默不语-个人CSDN博文目录

（这个文章本来是单文本摘要集合，但我觉得这么大的标题实在是毫无存在的意义，所以直接废物利用改成长文本摘要了。
如果之前的博文中有地方没有把链接和标题对应替换完，请联系我修改）
我之前几篇博文中分别写过相关的信息，具体如何解耦和修改我以后再慢慢来。

最近更新时间：2023.5.9
最早更新时间：2023.5.9

1. 抽取

1. (2019) Exploiting discourse-level segmentation for extractive summarization：discourse-level segmentation + adapted contextual representation model
   RNN或Bert
2. (2020) Discourse-Aware Neural Extractive Text Summarization

2. 抽取 - 具体咋做的还没看

1. (2021) Globalizing BERT-based Transformer Architectures for Long Document Summarization
2. (2021) Sliding Selector Network with Dynamic Memory for Extractive Summarization of Long Documents

3. 生成 - 分治

1. (2018) Deep Communicating Agents for Abstractive Summarization：强化学习。用agent分别处理每个subsection并进行信息交互
2. (2020) A Divide-and-Conquer Approach to the Summarization of Long Documents

4. 抽取→生成

1. (2021) Long Document Summarization in a Low Resource Setting using Pretrained Language Models
2. (2021) Long-Span Summarization via Local Attention and Content Selection

5. 生成 - 具体咋做的还没看

1. (2021) Hierarchical Learning for Generation with Long Source Sequences
2. (2021) Efficient attentions for long document summarization
3. (2022) Long Document Summarization with Top-Down and Bottom-Up Representation Inference