Modern History for Text Summarization

1. The Heat Change of Summarization Research (Updating...)


Fig.1 The number of accepted papers from ACL, EMNLP, NAACL over the past six years.

As shown in Fig.1, we can observe the influence of deep neural networks on the development of summarization. And next we will make brief descriptions of some key turning points:

2. Data Warehouse

Single-document Summarization
Multi-document Summarization
Multi-modal Summarization
New tasks for Summarization

3. Typical Research Problems

We have summarized and classified the research problems of existing work based on more than 100 papers in the last six years, to share them in the hope that more beginners can learn the most core things in this field more quickly.
Generating way
The number of source documents
New Datasets and Evaluations
Learning Methods
Multi-X
Defition of new tasks

4. Active Research Groups

China
North America & Europe
Company