A Scalable Global Model for Summarization

By D. Gillick, Benoit Favre. International Conference on Inductive Logic Programming, 2009

311 citations recorded by the source corpus.

We present an Integer Linear Program for exact inference under a maximum coverage model for automatic summarization. We compare our model, which operates at the sub-sentence or "concept-level, to a sentence-level model, previously solved with an ILP. Our model scales more efficiently to larger problems because it does not require a quadratic number of variables to address redundancy in pairs of selected sentences. We also show how to include sentence compression in the ILP formulation, which has the desirable property of performing compression and sentence selection simultaneously. The resulting system performs at least as well as the best systems participating in the recent Text Analysis Conference, as judged by a variety of automatic and manual content-based metrics.

Subfields: Artificial Intelligence

View Paper: https://www.semanticscholar.org/paper/8e2cd5369db9242574740e0d2739c755f8f61c92

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

Citations in the corpus

Reinforced Extractive Summarization with Question-Focused Rewards (2018, background): “This practice has shown success in earlier studies (Gillick and Favre, 2009).”
Learning to Score System Summaries for Better Content Selection Evaluation. (2017, methodology): “N-gram Coverage is inspired by the strong summarizer ICSI (Gillick and Favre, 2009).”
From Extractive to Abstractive Summarization: A Journey (2019, result): “We also compare the results with Submodular optimization based technique[43] and Integer linear programming based summarization[24], which are considered to be state of art techniques for sentence extraction[31].”
Objective Function Learning to Match Human Judgements for Optimization-Based Summarization (2018, methodology): “Baselines (1) ICSI (Gillick and Favre, 2009) is a global linear optimization approach that extracts a summary by solving a maximum coverage problem considering the most frequent bigrams in the source documents.”
Learning-Based Single-Document Summarization with Compression and Anaphoricity Constraints (2016, methodology): “According to ROUGE, our system outperforms a document prefix baseline, a bigram coverage baseline adapted from a strong multi-document system (Gillick and Favre, 2009), and a discourse-informed method from prior work (Yoshida et al.”
Jointly Learning to Extract and Compress (2011, methodology): “Similarly, Gillick and Favre (2009), though not learning weights, do a limited form of compression jointly with extraction.”
Neural Document Summarization by Jointly Learning to Score and Select Sentences (2018, unknown): “Gillick and Favre (2009) ﬁnd that using bigram features consistently yields better performance than unigrams or trigrams for ROUGE (Lin, 2004) measures.”
Using Local Knowledge Graph Construction to Scale Seq2Seq Models to Multi-Document Inputs (2019, background): “, 2008), extractive methods to select relevant sentences (Daumé III and Marcu, 2002; Gillick and Favre, 2009; Berg-Kirkpatrick et al., 2011; Di Fabbrizio et al., 2014; Bing et al., 2015; Cao et al., 2017) including maximal marginal relevance (Fabbri et al.”
Overcoming the Lack of Parallel Data in Sentence Compression (2013, methodology): “…et al., 2003; Zajic et al., 2007), (ii) supervised models trained on parallel data (Knight & Marcu, 2000; Turner & Charniak, 2005; McDonald, 2006; Gillick & Favre, 2009; Galanis & Androutsopoulos, 2010, inter alia) and (iii) unsupervised methods which make use of statistics collected from…”
Extractive Summarization as Text Matching (2020, methodology): “In addition, there is some work to solve extractive summarization from a semantic perspective before this paper, such as concept coverage (Gillick and Favre, 2009), reconstruction (Miao and Blun-som, 2016) and maximize semantic volume (Yo-gatama et al., 2015).”
Toward Abstractive Summarization Using Semantic Representations (2018, methodology): “ILPs and approximations have been used to encode compression and extraction (McDonald, 2007; Martins and Smith, 2009; Gillick and Favre, 2009; Berg-Kirkpatrick et al., 2011; Almeida and Martins, 2013; Li et al., 2014).”
The Role of Discourse Units in Near-Extractive Summarization (2016, methodology): “Finer-grained units such as n-grams are frequently used for quantifying content salience and redundancy prior to summarization over sentences (Filatova and Hatzivassiloglou, 2004; Thadani and McKeown, 2008; Gillick and Favre, 2009; Lin and Bilmes, 2011; Cao et al., 2015).”
Graph-based Neural Multi-Document Summarization (2017, methodology): “Then, they select summary-worthy sentences using a range of algorithms, such as graph centrality (Erkan and Radev, 2004), constraint optimization via Integer Linear Programming (Mc-Donald, 2007; Gillick and Favre, 2009; Li et al., 2013), or Support Vector Regression (Li et al., 2007) algorithms.”
Automatically Assessing Machine Summary Content Without a Gold Standard (2013, methodology): “To optimize a metric such as JS divergence exactly would be difﬁcult because the JS divergence score cannot be factored or divided among individual sentences, a necessary condition if the problem should be solved using an Integer Linear Program as in McDonald (2007) and Gillick and Favre (2009).”
Fast Abstractive Summarization with Reinforce-Selected Sentence Rewriting (2018, methodology): “…used Hidden Markov Models and rule-based systems (Jing and McKeown, 2000), statistical models based on parse trees (Knight and Marcu, 2000), and integer linear programming based methods (Martins and Smith, 2009; Gillick and Favre, 2009; Clarke and Lapata, 2010; Berg-Kirkpatrick et al., 2011).”
Reward Learning for Efficient Reinforcement Learning in Extractive Document Summarisation (2019, methodology): “For non-RL-based systems, we report ICSI [ Gillick and Favre, 2009 ] Ri-oux et al. , 2014 ] as an input-speciﬁc RL, and DeepTD as a cross-input RL. DeepTD is adapted from the DQN-based RL summariser [ Yao et al. , 2018 ] and is trained by taking U ∗ as the rewards (see § 4).”
The limits of automatic summarisation according to ROUGE (2017, methodology): “Of these, ICSISum (Gillick and Favre, 2009) is the only global summariser using an exact algorithm; it obtains the best ROUGE-2 score without supervision.”
Guiding Extractive Summarization with Question-Answering Rewards (2019, background): “…and optionally deleting unimportant sentence constituents to form a summary (Knight and Marcu, 2002; Radev et al., 2004; Zajic et al., 2007; Martins and Smith, 2009; Gillick and Favre, 2009; Lin and Bilmes, 2010; Wang et al., 2013; Li et al., 2013, 2014; Hong et al., 2014; Yogatama et al., 2015).”
Long story short - Global unsupervised models for keyphrase based meeting summarization (2010, methodology): “A comprehensive comparison of this model against McDonald’s in term of scalability can be found in (Gillick and Favre, 2009).”
Min-Cost Flow Routing for Evidence Assembly in Long Multimodal Documents (2026, unknown): “Evidence selection has also been formulated using integer linear programming (McDonald, 2007; Gillick & Favre, 2009), submodular maximization (Lin & Bilmes, 2010, 2011), MMR (Carbonell & Goldstein, 1998), and determinantal point processes (Kulesza & Taskar, 2012; Cho et al., 2019).”