E14-1075 experiments on two datasets from the Document Understanding Conference ( DUC ) ( Over et
C04-1129 30 document sets from the 2003 Document Understanding Conference ( see a36 3.1 for
D08-1032 Setup We used the main task of Document Understanding Conference ( DUC ) 2007 for evaluation
D13-1069 conveniently . A typical instance is the Document Understanding Conferences ( DUC ) 1 , which
E06-1039 well-formedness questions used at the 2005 Document Understanding Conference ( DUC ) ( Nat , 2005a
D14-1075 length match those set by the Document Understanding Conferences associated with the
D08-1080 summarization framework of the Document Understanding Conference have confirmed that
E09-1062 was based on results from the Document Understanding Conference ( DUC ) evaluations
C04-1129 Rouge ( a36 3.2 ) . 3.1 Data The Document Understanding Conference ( DUC ) has been run
C04-1077 al. , 2002 ) took place and the Document Understanding Conference ( DUC ) 1 has been
D15-1011 experiment on six datasets from the Document Understanding Conference ( DUC ) and the Text
C94-2108 can then be used to inform other document understanding techniques , including noun phrase
C04-1057 multidocument summarization during the Document Understanding Conference ( DUC ) , organized
E09-1048 evaluation campaigns , such as NIST 's Document Understanding Conferences ( DUC ) , impose
D14-1075 query focused tasks run in the Document Understanding Conference ( DUC ) and Text Analysis
D15-1222 for more than a decade in the Document Understanding Conference ( DUC ) , and subsequently
D13-1068 problem to be solved . Although many document understanding systems have been proposed ,
D09-1044 summarisation data from the 2001 Document Understanding Conference ( DUC ) ,8 which is
D15-1223 experiments on three corpora from the Document Understanding Conference ( DUC ) : 2002 , 2004
E14-1067 human written summaries from the Document Understanding Conference ( DUC3 ) evalua tion
hide detail