Share this page:

Document-level Event Extraction with Efficient End-to-end Learning of Cross-event Dependencies

Kung-Hsiang Huang and Nanyun Peng, in The 3rd Workshop on Narrative Understanding (NAACL 2021), 2021.

Download the full text


Abstract

Fully understanding narratives often requires identifying events in the context of whole documents and modeling the event relations. However, document-level event extraction is a challenging task as it requires the extraction of event and entity coreference, and capturing arguments that span across different sentences. Existing works on event extraction usually confine on extracting events from single sentences, which fail to capture the relationships between the event mentions at the scale of a document, as well as the event arguments that appear in a different sentence than the event trigger. In this paper, we propose an end-to-end model leveraging Deep Value Networks (DVN), a structured prediction algorithm, to efficiently capture cross-event dependencies for document-level event extraction. Experimental results show that our approach achieves comparable performance to CRF-based models on ACE05, while enjoys significantly higher computational efficiency.


Bib Entry

@inproceedings{huang2021document,
  title = {Document-level Event Extraction with Efficient End-to-end Learning of Cross-event Dependencies},
  author = {Huang, Kung-Hsiang and Peng, Nanyun},
  booktitle = {The 3rd Workshop on Narrative Understanding (NAACL 2021)},
  year = {2021}
}

Related Publications

  • Document-level Event Extraction with Efficient End-to-end Learning of Cross-event Dependencies

    Kung-Hsiang Huang and Nanyun Peng, in The 3rd Workshop on Narrative Understanding (NAACL 2021), 2021.
    Full Text Abstract BibTeX Details
    Fully understanding narratives often requires identifying events in the context of whole documents and modeling the event relations. However, document-level event extraction is a challenging task as it requires the extraction of event and entity coreference, and capturing arguments that span across different sentences. Existing works on event extraction usually confine on extracting events from single sentences, which fail to capture the relationships between the event mentions at the scale of a document, as well as the event arguments that appear in a different sentence than the event trigger. In this paper, we propose an end-to-end model leveraging Deep Value Networks (DVN), a structured prediction algorithm, to efficiently capture cross-event dependencies for document-level event extraction. Experimental results show that our approach achieves comparable performance to CRF-based models on ACE05, while enjoys significantly higher computational efficiency.
    @inproceedings{huang2021document,
      title = {Document-level Event Extraction with Efficient End-to-end Learning of Cross-event Dependencies},
      author = {Huang, Kung-Hsiang and Peng, Nanyun},
      booktitle = {The 3rd Workshop on Narrative Understanding (NAACL 2021)},
      year = {2021}
    }
    
    Details
  • Biomedical Event Extraction with Hierarchical Knowledge Graphs

    Kung-Hsiang Huang, Mu Yang, and Nanyun Peng, in the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)-Findings, short, 2020.
    Full Text Slides Code Abstract BibTeX Details
    Biomedical event extraction is critical in understanding biomolecular interactions described in scientific corpus. One of the main challenges is to identify nested structured events that are associated with non-indicative trigger words. We propose to incorporate domain knowledge from Unified Medical Language System (UMLS) to a pre-trained language model via a hierarchical graph representation encoded by a proposed Graph Edgeconditioned Attention Networks (GEANet). To better recognize the trigger words, each sentence is first grounded to a sentence graph based on a jointly modeled hierarchical knowledge graph from UMLS. The grounded graphs are then propagated by GEANet, a novel graph neural networks for enhanced capabilities in inferring complex events. On BioNLP 2011 GENIA Event Extraction task, our approach achieved 1.41% F1 and 3.19% F1 improvements on all events and complex events, respectively. Ablation studies confirm the importance of GEANet and hierarchical KG.
    @inproceedings{huang2020event,
      title = {Biomedical Event Extraction with Hierarchical Knowledge Graphs},
      author = {Huang, Kung-Hsiang and Yang, Mu and Peng, Nanyun},
      booktitle = {the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)-Findings, short},
      slideslive_id = {38940169},
      year = {2020}
    }
    
    Details