Automatic Text Summarization of Newswire: Lessons Learned from the Document Understanding Conference

By Ani Nenkova. 2005

175 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Natural Language Processing

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Natural Language Processing

Citations in the corpus

Repairing the Cracked Foundation: A Survey of Obstacles in Evaluation Practices for Generated Text (2022, background): “fessionally written article is a strong baseline, as shown by Nenkova (2005) who analyze DUC-2001 data and note that “only one system significantly”
Don’t Give Me the Details, Just the Summary! Topic-Aware Convolutional Neural Networks for Extreme Summarization (2018, methodology): “LEAD is often used as a strong lower bound for news summarization (Nenkova, 2005) and creates a summary by selecting the first”
On Faithfulness and Factuality in Abstractive Summarization (2020, background): “Following the Document Understanding Conference (DUC; Dang, 2005), a majority of work has focused on evaluating the content and the linguistic quality of summaries (Nenkova, 2005).”
Content Selection in Deep Learning Models of Summarization (2018, background): “Document Shuffling Sentence position is a well known and powerful feature for news summarization (Hong and Nenkova, 2014), owing to the intentional lead bias in the news article writing10; it also explains the difficulty in beating the lead baseline for single-document summarization (Nenkova, 2005; ”
Detecting Information-Dense Texts in Multiple News Domains (2014, background): “The results from largescale evaluations have confirmed that human summaries easily outperform the baseline of selecting the beginning of the article as a summary but automatic approaches to summarization cannot perform significantly better than this baseline on manual evaluation metrics (Nenkova 200”
Enhancing Single-Document Summarization by Combining RankNet and Third-Party Sources (2007, methodology): “In recent years, there has been a decline in studies on automatic single-document summarization, in part because the DUC task was dropped, and in part because the task of single-document extracts may be counterintuitively more difﬁcult than multi-document summarization (Nenkova, 2005).”
Automatic Generation of Story Highlights (2010, unknown): “This “lead” baseline may err on the side of verbosity but at least will be grammatical, and it has indeed proved extremely hard to outperform by more sophisticated methods (Nenkova, 2005).”
MultiLing 2015: Multilingual Summarization of Single and Multi-Documents, On-line Fora, and Call-center Conversations (2015, background): “Historically such tasks have predominantly focused on English news documents, see for example Nenkova (2005).”
How well do you know your summarization datasets? (2021, unknown): “S UMM A BSTRACCTIVE (Liu and Lapata, 2019), (7) P RE S UMM T RANSFORMER (Liu and Lapata, 2019), (8) LEAD (Nenkova, 2005), (9) E XT O RA - CLE (Nallapati et al., 2017)”
A Survey on Automatic Text Summarization (2007, background): “This extremely strong baseline has been analyzed by Nenkova (2005) and corresponds to the selection of the first n sentences of a newswire article.”