Extracting Relational Facts by an End-to-End Neural Model with Copy Mechanism

By Xiangrong Zeng, Daojian Zeng, Shizhu He, Kang Liu, Jun Zhao. Annual Meeting of the Association for Computational Linguistics, 2018

476 citations recorded by the source corpus.

The relational facts in sentences are often complicated. Different relational triplets may have overlaps in a sentence. We divided the sentences into three types according to triplet overlap degree, including Normal, EntityPairOverlap and SingleEntiyOverlap. Existing methods mainly focus on Normal class and fail to extract relational triplets precisely. In this paper, we propose an end-to-end model based on sequence-to-sequence learning with copy mechanism, which can jointly extract relational facts from sentences of any of these classes. We adopt two different strategies in decoding process: employing only one united decoder or applying multiple separated decoders. We test our models in two public datasets and our model outperform the baseline method significantly.

Subfields: Artificial Intelligence

View Paper: https://www.semanticscholar.org/paper/416fcbf760e41857b7530826a8803c34de2ab077

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

Unified Structure Generation for Universal Information Extraction (2022, methodology): “…relevant to our work are generation-based IE methods, which generate text spans via tagging (Straková et al., 2019; Ma et al., 2019), index pointer (Ren et al., 2021; Yan et al., 2021b) or copy mechanism (Zeng et al., 2018), and these methods usually employ speciﬁc classiﬁers to represent labels.”
Structured Prediction as Translation between Augmented Natural Languages (2021, methodology): “• The NYT dataset (Zeng et al., 2018) is based on the New York Times corpus and was automatically labeled with distant supervision by Riedel et al. (2010).”
Text2Event: Controllable Sequence-to-Structure Generation for End-to-end Event Extraction (2021, methodology): “, 2019) or copy/pointer mechanism (Zeng et al., 2018; Du et al., 2021), TEXT2EVENT directly generate event schemas and text spans to form event records via constrained decoding (Cao et al.”
BioGPT: Generative Pre-trained Transformer for Biomedical Text Generation and Mining (2022, methodology): “Text generation methods reframe the task as a sequence-to-sequence learning task, by taking the text as the input sequence and the triplet as the target sequence and employing an encoder-decoder network to learn to generate the triplet from the text [42, 43, 14, 24, 25].”
GraphRel: Modeling Text as Relational Graphs for Joint Entity and Relation Extraction (2019, methodology): “Zeng et al. (2018) then propose an end-to-end sequence-to-sequence model for relation extraction.”
Entity-Relation Extraction as Multi-Turn Question Answering (2019, methodology): “Wang et al. (2016a) extract relations using multi-level attention CNNs. Zeng et al. (2018) proposed a new framework that uses sequence-to-sequence models to generate entity-relation triples, naturally combining entity detection and relation detection.”
TPLinker: Single-stage Joint Extraction of Entities and Relations Through Token Pair Linking (2020, methodology): “Following Zeng et al. (2018), to further study Table 1: Statistics of the datasets in our experiments.”
A Novel Cascade Binary Tagging Framework for Relational Triple Extraction (2019, methodology): “…and the heads of both subject and object are all correct (Li and Ji, 2014; Miwa and Bansal, 2016; Katiyar and Cardie, 2017; Zheng et al., 2017; Zeng et al., 2018; Takanobu These two versions have been selectively adopted and preprocessed in many different ways among various previous works,…”
Retrieval And Structuring Augmented Generation with Large Language Models (2025, unknown): “RoSTER [167] addressed incomplete labels using noise-robust learning.”