Tailoring outputs of large language models, such as ChatGPT, to specific...
Discourse processing suffers from data sparsity, especially for dialogue...
Despite the success of recent abstractive summarizers on automatic evalu...
The transformer multi-head self-attention mechanism has been thoroughly
...
Recently proposed pre-trained generation models achieve strong performan...
Transformers are the dominant architecture in NLP, but their training an...
Aiming for a better integration of data-driven and linguistically-inspir...
In news articles the lead bias is a common phenomenon that usually domin...
Previous work indicates that discourse information benefits summarizatio...
The multi-head self-attention of popular transformer models is widely us...
Our analysis of large summarization datasets indicates that redundancy i...
An essential prerequisite for unleashing the potential of supervised dee...
In this paper, we propose a novel neural single document extractive
summ...