Skip to Main content Skip to Navigation
Conference papers

RL extraction of syntax-based chunks for sentence compression

Hoa Le 1 Christophe Cerisara 1 Claire Gardent 1
1 SYNALP - Natural Language Processing : representations, inference and semantics
LORIA - NLPKD - Department of Natural Language Processing & Knowledge Discovery
Abstract : Sentence compression involves selecting key information present in the input and rewriting this information into a short, coherent text. While dependency parses have often been used for this purpose, we propose to exploit such syntactic information within a modern reinforcement learning-based extraction model. Furthermore, compared to other approaches that include syntactic features into deep learning models, we design a model that has better explainability properties and is flexible enough to support various shallow syntactic parsing modules. More specifically, we linearize the syntactic tree into the form of overlapping text segments, which are then selected with reinforcement learning and regenerated into a compressed form. Hence, despite relying on extractive components, our model is also able to handle abstractive summarization. We explore different ways of selecting subtrees from the dependency structure of the input sentence and compare the results of various models on the Gigaword corpus.
Complete list of metadata

Cited literature [20 references]  Display  Hide  Download
Contributor : Christophe Cerisara <>
Submitted on : Tuesday, October 20, 2020 - 5:57:40 PM
Last modification on : Thursday, March 11, 2021 - 2:26:02 PM


Files produced by the author(s)


  • HAL Id : hal-02323821, version 2


Hoa Le, Christophe Cerisara, Claire Gardent. RL extraction of syntax-based chunks for sentence compression. ICANN 2019, Sep 2019, Munich, Germany. pp.337-347. ⟨hal-02323821v2⟩



Record views


Files downloads