Diversity and Quality: Comparing Decoding Methods with PEGASUS for Text Summarization

Simple item page
dc.contributor陳柏琳zh_TW
dc.contributorChen, Berlinen_US
dc.contributor.author唐科南zh_TW
dc.contributor.authorThompson, Keenan Nathanielen_US
dc.date.accessioned2022-06-08T02:43:38Z
dc.date.available2021-10-29
dc.date.available2022-06-08T02:43:38Z
dc.date.issued2021
dc.description.abstractnonezh_TW
dc.description.abstractThis thesis offers three major contributions: (1) It considers a number of diverse decoding methods to address degenerate repetition in model output text and investigates what can be done to mitigate the loss in summary quality associated with the use of such methods. (2) It provides evidence that measure of textual lexical diversity (MTLD) is as viable tool as perplexity is for comparing text diversity in this context. (3) It presents a detailed analysis of the strengths and shortcomings of ROUGE, particularly in regard to abstractive summarization. To explore these issues the work analyzes the results of experiments run on the CNN/DailyMail dataset with the PEGASUS model.en_US
dc.description.sponsorship資訊工程學系zh_TW
dc.identifier60847093S-40679
dc.identifier.urihttps://etds.lib.ntnu.edu.tw/thesis/detail/8a988c2eaa76dfbc4f7d610b1e56fba9/
dc.identifier.urihttp://rportal.lib.ntnu.edu.tw/handle/20.500.12235/117349
dc.language英文
dc.subjectnonezh_TW
dc.subjectsummarizationen_US
dc.subjectdiverse decodingen_US
dc.subjectPEGASUSen_US
dc.subjectROUGEen_US
dc.subjectlexical diversityen_US
dc.titleDiversity and Quality: Comparing Decoding Methods with PEGASUS for Text Summarizationzh_TW
dc.titleDiversity and Quality: Comparing Decoding Methods with PEGASUS for Text Summarizationen_US
dc.type學術論文

Files

Original bundle

Now showing 1 - 1 of 1
No Thumbnail Available
Name:
60847093S-40679.pdf
Size:
377.71 KB
Format:
Adobe Portable Document Format
Description:
學術論文
Download

Collections

學位論文