Abstract
Most existing event extraction works mainly focus on extracting events from one sentence. However, in real-world applications, arguments of one event may scatter across sentences and multiple events may co-occur in one document. Thus, these scenarios require document-level event extraction (DEE), which aims to extract events and their arguments across sentences from a document. Previous works cast DEE as a two-step paradigm: sentence-level event extraction (SEE) to document-level event fusion. However, this paradigm lacks integrating document-level information for SEE and suffers from the inherent limitations of error propagation. In this article, we propose a multi-turn and multi-granularity reader for DEE that can extract events from the document directly without the stage of preliminary SEE. Specifically, we propose a new paradigm of DEE by formulating it as a machine reading comprehension task (i.e., the extraction of event arguments is transformed to identify the answer span from the document). Beyond the framework of machine reading comprehension, we introduce a multi-turn and multi-granularity reader to capture the dependencies between arguments explicitly and model long texts effectively. The empirical results demonstrate that our method achieves superior performance on the MUC-4 and the ChFinAnn datasets.
- [1] . 2014. Modeling biological processes for reading comprehension. In Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP’14). 1499–1510.Google Scholar
Cross Ref
- [2] . 2018. Biomedical event extraction using convolutional neural networks and dependency parsing. In Proceedings of the BioNLP 2018 Workshop. 98–108. Google Scholar
Cross Ref
- [3] . 2019. Rapid customization for event extraction. In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics: System Demonstrations. 31–36. Google Scholar
Cross Ref
- [4] . 2020. Reconstructing event regions for event extraction via graph attention networks. In Proceedings of the 1st Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics and the 10th International Joint Conference on Natural Language Processing. 811–820. https://www.aclweb.org/anthology/2020.aacl-main.81.Google Scholar
- [5] . 2015. Event extraction via dynamic multi-pooling convolutional neural networks. In Proceedings of the 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Conference on Natural Language Processing (Volume 1: Long Papers). 167–176.Google Scholar
Cross Ref
- [6] . 2019. BERT: Pre-training of deep bidirectional transformers for language understanding. In Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long and Short Papers). 4171–4186. Google Scholar
Cross Ref
- [7] . 2004. The automatic content extraction (ACE) program-tasks, data, and evaluation. In Proceedings of the 4th International Conference on Language Resources and Evaluation (LREC’04).Google Scholar
- [8] . 2020. Document-level event role filler extraction using multi-granularity contextualized encoding. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics. 8010–8020. Google Scholar
Cross Ref
- [9] . 2020. Event extraction by answering (almost) natural questions. In Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP’20). 671–683. Google Scholar
Cross Ref
- [10] . 2020. Document-level event-based extraction using generative template-filling transformers. arXiv preprint arXiv:2008.09249 (2020).Google Scholar
- [11] . 2011. Peeling back the layers: Detecting event role fillers in secondary contexts. In Proceedings of the 49th Annual Meeting of the Association for Computational Linguistics: Human Language Technologies (Volume 1). 1137–1147.Google Scholar
- [12] . 2012. Bootstrapped training of event extraction classifiers. In Proceedings of the 13th Conference of the European Chapter of the Association for Computational Linguistics. 286–295.Google Scholar
Digital Library
- [13] . 2011. Knowledge base population: Successful approaches and challenges. In Proceedings of the 49th Annual Meeting of the Association for Computational Linguistics: Human Language Technologies. 1148–1158.Google Scholar
Digital Library
- [14] . 2017. TriviaQA: A large scale distantly supervised challenge dataset for reading comprehension. In Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 1601–1611. Google Scholar
Cross Ref
- [15] . 2014. ADAM: A method for stochastic optimization. arXiv preprint arXiv:1412.6980 (2014).Google Scholar
- [16] . 2017. RACE: Large-scale ReAding comprehension dataset from examinations. In Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing. 785–794. Google Scholar
Cross Ref
- [17] . 2017. Zero-shot relation extraction via reading comprehension. In Proceedings of the 21st Conference on Computational Natural Language Learning (CoNLL’17). 333–342. Google Scholar
Cross Ref
- [18] . 2020. Event extraction as multi-turn question answering. In Findings of the Association for Computational Linguistics: EMNLP 2020. 829–838. Google Scholar
Cross Ref
- [19] . 2020. Connecting the dots: Event graph schema induction with path language modeling. In Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP’20). 684–695. Google Scholar
Cross Ref
- [20] . 2020. A unified MRC framework for named entity recognition. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics. 5849–5859. Google Scholar
Cross Ref
- [21] . 2019. Entity-relation extraction as multi-turn question answering. In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics. 1340–1350. Google Scholar
Cross Ref
- [22] . 2020. Event extraction as machine reading comprehension. In Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP’20). 1641–1651. Google Scholar
Cross Ref
- [23] . 2018. The natural language decathlon: Multitask learning as question answering. arXiv preprint arXiv:1806.08730 (2018).Google Scholar
- [24] . 1992. Fourth Message Understanding Conference (MUC-4): Proceedings of a Conference Held in McLean, Virginia, June 16-18, 1992. Morgan Kaufman, San Mateo, CA. https://www.aclweb.org/anthology/M92-1000.Google Scholar
- [25] . 2016. Joint event extraction via recurrent neural networks. In Proceedings of the 2016 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies. 300–309.Google Scholar
Cross Ref
- [26] . 2009. A unified model of phrasal and sentential evidence for information extraction. In Proceedings of the 2009 Conference on Empirical Methods in Natural Language Processing (Volume 1). 151–160.Google Scholar
Digital Library
- [27] . 2016. SQuAD: 100,000+ questions for machine comprehension of text. In Proceedings of the 2016 Conference on Empirical Methods in Natural Language Processing. 2383–2392. Google Scholar
Cross Ref
- [28] . 2020. Biomedical event extraction as sequence labeling. In Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP’20). 5357–5367. Google Scholar
Cross Ref
- [29] . 2016. Bidirectional attention flow for machine comprehension. arXiv preprint arXiv:1611.01603 (2016).Google Scholar
- [30] . 2017. ReasoNet: Learning to stop reading in machine comprehension. In Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. 1047–1055.Google Scholar
Digital Library
- [31] . 2017. Attention is all you need. In Advances in Neural Information Processing Systems. 5998–6008.Google Scholar
- [32] . 2016. Machine comprehension using match-LSTM and answer pointer. arXiv preprint arXiv:1608.07905 (2016).Google Scholar
- [33] . 2019. HMEAE: Hierarchical modular event argument extraction. In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP’19). 5781–5787.Google Scholar
Cross Ref
- [34] . 2019. Multi-passage BERT: A globally normalized BERT model for open-domain question answering. arXiv preprint arXiv:1908.08167 (2019).Google Scholar
- [35] . 2021. Document-level event extraction via heterogeneous graph-based interaction model with a tracker. In Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers). 3533–3546. Google Scholar
Cross Ref
- [36] . 2018. DCFEE: A document-level Chinese financial event extraction system based on automatically labeled training data. In Proceedings of ACL 2018: System Demonstrations. 50–55.Google Scholar
Cross Ref
- [37] . 2019. Exploring pre-trained language models for event extraction and generation. In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics. 5284–5294.Google Scholar
Cross Ref
- [38] . 2018. HotpotQA: A dataset for diverse, explainable multi-hop question answering. In Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing. 2369–2380. Google Scholar
Cross Ref
- [39] . 2020. Document modeling with graph attention networks for multi-grained machine reading comprehension. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics. 6708–6718. Google Scholar
Cross Ref
- [40] . 2019. Doc2EDAG: An end-to-end document-level framework for Chinese financial event extraction. arXiv preprint arXiv:1904.07535 (2019).Google Scholar
Index Terms
Multi-Turn and Multi-Granularity Reader for Document-Level Event Extraction
Recommendations
Explainable document-level event extraction via back-tracing to sentence-level event clues
AbstractDocument-level event extraction (DEE) aims to extract events directly from a document. Current state-of-the-art methods follow the entity-centric framework, capturing sentence-level event clues into entity instance representations for ...
Graphical abstractDisplay Omitted
Highlights- An explainable document-level event extraction framework ExpDEE is proposed.
- ...
Document-Level Multi-event Extraction via Event Ontology Guiding
Knowledge Science, Engineering and ManagementAbstractDocument-level Event Extraction (DEE) aims to extract event information from a whole document, in which extracting multiple events is a fundamental challenge. Previous works struggle to handle the Document-level Multi-Event Extraction (DMEE) due ...
DuEE-Fin: A Large-Scale Dataset for Document-Level Event Extraction
Natural Language Processing and Chinese ComputingAbstractTo tackle the data scarcity problem of document-level event extraction, we come up with a large-scale benchmark, DuEE-Fin, which consists of 15,000+ events categorized into 13 event types, and 81,000+ event arguments mapped in 92 argument roles. ...






Comments