Improving faithfulness in abstractive text summarization with EDUs using BART

dc.contributor.authorDelpisheh, Narjes
dc.contributor.authorUniversity of Lethbridge. Faculty of Arts and Science
dc.contributor.supervisorChali, Yllias
dc.date.accessioned2023-05-24T17:09:29Z
dc.date.available2023-05-24T17:09:29Z
dc.date.issued2023-04-27
dc.degree.levelMasters
dc.description.abstractAbstractive summarization aims to reproduce the essential information of a source document in a summary by using the summarizer's own words. Although this approach is more similar to how humans summarize, it is more challenging to automate as it requires a complete understanding of natural language. However, the development of deep learning approaches, such as the sequence-to-sequence model with an attention-based mechanism, and the availability of pre-trained language models have led to improved performance in summarization systems. Nonetheless, abstractive summarization still suffers from issues such as hallucination and unfaithfulness. To address these issues, we propose an approach that utilizes a guidance signal using important Elementary Discourse Units (EDUs). We compare our work with previous guided summarization and two other summarization models that enhanced the faithfulness of the summary. Our approach was tested on CNN/Daily Mail dataset, and results showed an improvement in both truthfulness and good quantity coverage of the source document.
dc.identifier.urihttps://hdl.handle.net/10133/6499
dc.language.isoen_US
dc.proquest.subject0984
dc.proquestyesYes
dc.publisherLethbridge, Alta. : University of Lethbridge, Dept. of Mathematics and Computer Science
dc.publisher.departmentDepartment of Mathematics and Computer Science
dc.publisher.facultyArts and Science
dc.relation.ispartofseriesThesis (University of Lethbridge. Faculty of Arts and Science)
dc.subjectFaithfulness
dc.subjectAbstractive summarization
dc.subjectElementary discourse units
dc.subjectEDU
dc.subjectHallucination
dc.subjectText summarization
dc.subjectNatural language
dc.subject.lcshNatural language processing (Computer science)
dc.subject.lcshHuman-computer interaction
dc.subject.lcshSemantic computing
dc.subject.lcshDissertations, Academic
dc.titleImproving faithfulness in abstractive text summarization with EDUs using BART
dc.typeThesis
Files
Original bundle
Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
NARJES_DELPISHEH_MSC_2023.pdf
Size:
939.67 KB
Format:
Adobe Portable Document Format
Description:
License bundle
Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
license.txt
Size:
3.25 KB
Format:
Item-specific license agreed upon to submission
Description: