Login Register
Back to top
Loading...
Loading...

Guided Research Aditi Arora

Last modified Aug 11, 2023
   No tags assigned

Exploring Domain Adaptation Techniques for Abstractive Text Summarization

 

Introduction:

 

In recent years, with the growth of the Internet, the amount of text data from various sources has grown exponentially. This text volume is an invaluable source of information and knowledge that needs to be effectively and quickly summarized to be useful. Automatic text summarization is a subtask of Natural Language Generation (NLG) that aims to produce compressed summaries of text data. Abstractive text summarization methods are used to generate meaningful summaries that reflect the core idea of the text. Abstractive summarization is challenging because it requires a large amount of human-annotated data. In recent years, neural abstractive text summarization using sequence-to-sequence models (seq2seq) has gained much popularity. Although transformer-based models have provided promising results in this area, the lack of labeled data for domain-specific texts remains one of the significant challenges. This work provides a detailed and comprehensive overview of different domain adaptation techniques utilized for abstractive text summarization.
 
Research Questions:
 
Question 1: 

What are the different types of models being used for domain adaptation in abstractive text summarization?

Question 2:

What are the different types of domain adaptation techniques used for abstractive text summarization?

Question 3:

What is the criteria for classifying the various domain adaptation techniques found in the literature?

 

 

 
 

 

 

 

 

 

Incoming references

Files and Subpages

Name Type Size Last Modification Last Editor
Aditi Arora GR Final Presentation May.pdf
File
2,86 MB 11.08.2023
Aditi Arora GR-Kickoff.pptx
File
1,60 MB 11.08.2023
Save Cancel