2007
DOI: 10.1016/j.ipm.2007.01.019
|View full text |Cite
|
Sign up to set email alerts
|

DUC in context

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

0
94
0

Year Published

2009
2009
2017
2017

Publication Types

Select...
6
3

Relationship

0
9

Authors

Journals

citations
Cited by 158 publications
(94 citation statements)
references
References 2 publications
0
94
0
Order By: Relevance
“…The wide adoption of such automatic measures is understandable because they are convenient and greatly reduce the complexity of evaluations. ROUGE scores also correlate well with manual evaluations of content based on comparison with a single model summary, as used in the early editions of the Document Understanding Conferences (Over et al, 2007).…”
Section: Introductionmentioning
confidence: 76%
“…The wide adoption of such automatic measures is understandable because they are convenient and greatly reduce the complexity of evaluations. ROUGE scores also correlate well with manual evaluations of content based on comparison with a single model summary, as used in the early editions of the Document Understanding Conferences (Over et al, 2007).…”
Section: Introductionmentioning
confidence: 76%
“…We take advantage of two currently available resources: archival data from the Document Understanding Conferences (DUC) (Over et al, 2007) and the New York Times (NYT) corpus (https://catalog.ldc.upenn.edu/ LDC2008T19).…”
Section: Corporamentioning
confidence: 99%
“…Specifically, we demonstrate that the content-density detector is able to recognize when an automatic summary of a single news article is better than the lead of the article. This is an important open problem in summarization, where the lead paragraph baseline is very strong and few (Over, Dang, & Harman, 2007). Moreover, all of the proof-of-concept experiments from the previous section were performed on data drawn from the NYT.…”
Section: Recognizing Better Summariesmentioning
confidence: 99%