BBN/UMD at DUC-2004: Topiary

By David M. Zajic, Bonnie J. Dorr, Richard M. Schwartz. 2004

101 citations recorded by the source corpus.

How researchers see this paper

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

In Natural Language Processing

Citations in the corpus

Abstractive Text Summarization using Sequence-to-sequence RNNs and Beyond (2016, methodology): “The best performing system on the DUC-2004 task, called TOPIARY (Zajic et al., 2004), used a combination of linguistically motivated compression techniques, and an unsupervised topic detection algorithm that appends keywords extracted from the article onto the compressed output.”
A Neural Attention Model for Abstractive Sentence Summarization (2015, methodology): “T OPIARY merges a compression system using linguistically-motivated transformations of the input (Dorr et al., 2003) with an unsupervised topic detection (UTD) algorithm that appends key phrases from the full article onto the compressed output.”
Controlling Output Length in Neural Encoder-Decoders (2016, methodology): “Traditional approaches to this task focus on word deletion using rule-based (Dorr et al., 2003; Zajic et al., 2004) or statistical (Woodsend et al., 2010; Galanis and Androutsopoulos, 2010; Filippova and Strube, 2008; Filippova and Altun, 2013; Filip-pova et al., 2015) methods.”
Fast Abstractive Summarization with Reinforce-Selected Sentence Rewriting (2018, neutral): “The latter abstractive approach involves rewriting the summary (Banko et al., 2000; Zajic et al., 2004), and has seen substantial recent gains due to neural sequence-to-sequence models (Chopra et al., 2016; Nallap-ati et al., 2016; See et al., 2017; Paulus et al., 2018).”
Neural Headline Generation with Minimum Risk Training (2016, background): “• TOPIARY [21] is the winner system of DUC2004 Task1.”
Global Encoding for Abstractive Summarization (2018, methodology): “Researchers developed many statistical methods and linguistic-rule-based methods to study automatic summarization (Banko et al., 2000; Dorr et al., 2003; Zajic et al., 2004; Cohn and Lapata, 2008).”
HEADS: Headline Generation as Sequence Prediction Using an Abstract Feature-Rich Space (2015, methodology): “For performing an automatic evaluation of the head-lines generated by our system we follow the path taken by related work such as Zajic et al. (2004) and use a subset of the ROUGE metrics for comparing candidate headlines with reference ones, which have been proven to strongly correlate with human…”
The Factual Inconsistency Problem in Abstractive Text Summarization: A Survey (2021, unknown): “Conventional abstractive summarization methods usually extract some keywords from the source document, and then reorder and perform linguistically motivated transformations to these keywords [ Zajic et al. , 2004 ] .”
Abstract Text Summarization with a Convolutional Seq2seq Model (2019, methodology): “TOPIARY [16] utilized a combination of compression techniques motivated by linguistic, and an unsupervised topic detection method which appends keywords extracted from the story to the compressed output.”
Corpus-Guided Sentence Generation of Natural Images (2011, result): “Given this fact, our proposed approach performance is comparable to other state of the art summarization work in the literature [Bonnie and Dorr, 2004].”
Automatic Caption Generation for News Images (2013, background): “Headline generation is such a task that generates a very short title-like summary for a given document (Witbrock and Mittal, 1999; Banko et al., 2000; Jin and Hauptmann, 2001a,b, 2002; Zajic et al., 2002; Zhou and Hovy, 2003; Dorr et al., 2003; Bonnie et al., 2004).”
Neural Headline Generation with Sentence-wise Optimization (2016, background): “The best system on DUC-2004, TOPIARY (Zajic et al., 2004) com-”
Title Generation with Quasi-Synchronous Grammar (2010, methodology): “The compressed sentence may also be “padded” with important content words or phrases to ensure that the topic of the document is covered (Zajic et al., 2004).”