Elamur Research
☰
home / papers / 2021 / Document-Level Event Argument Extraction by Conditional Generation

Document-Level Event Argument Extraction by Conditional Generation

Written for the Research library (claude-opus-5-5) from the paper text. Original — aclanthology.org/2021.naacl-main.69.pdf

Summary

Most event extraction systems work within single sentences. In real documents, however, an event's participants are often mentioned in neighbouring sentences or only as pronouns, so sentence-level extraction gives incomplete or uninformative results. Li, Ji and Han reframe document-level argument extraction as conditional text generation. An encoder-decoder model (BART; T5 is also discussed) receives two inputs: the event-type template from the ontology, with <arg> slots, and the document, with the target trigger marked by <tgr> tokens. The model then writes out the completed template, and argument spans are read from it. Decoding can only copy tokens from the input, which prevents invented arguments. Clarification statements reintroduce entity-type constraints. The model needs no separate entity recognition or coreference step and handles empty or multi-filler roles naturally. The authors also release WIKIEVENTS, a document-level benchmark annotated fully for events and coreference, and propose informative argument extraction, which asks for the most descriptive mention, such as a name rather than a pronoun. The model beats CRF- and QA-based baselines clearly. Because it is template-driven, it also transfers to unseen event types.

Key points

  • F1 gains over the next best system:
    • RAMS: +7.6 absolute F1.
    • WIKIEVENTS: +5.7 absolute F1.
    • Informative argument extraction: +9.3 F1 over the strongest baseline.
  • Only 34.5% of arguments found in the trigger's own sentence count as informative mentions. This motivates resolving arguments across the whole document.
  • The first end-to-end zero-shot event extraction pipeline combines a keyword-based trigger model (TapKey) with BART-Gen. Trained on the 10 most frequent of the 33 ACE types, it scores:
    • Trigger classification: 69.23 TC F1, versus 71.13 fully supervised (about 97%).
    • Argument classification: 44.19 AC Head F1, versus 53.71 fully supervised (about 82%).
  • ACE has no templates of its own, so templates were borrowed from the RAMS AIDA and KAIROS ontologies.
  • Code and data are available at github.com/raspberryice/gen-arg.
  • Within the SM-KBP2020 system, the model discovered 53% more arguments, especially ones far from the trigger.

Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.

Fields: Natural language processing

Cited by

178 papers of our library cite it (of the 140673 whose bibliographies we read); 3 of them have a page here, 175 not yet.

Cite

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.18653/v1/2021.naacl-main.69

@inproceedings{li2021document,
  title = {{Document-Level Event Argument Extraction by Conditional Generation}},
  author = {Li, Sha and Ji, Heng and Han, Jiawei},
  booktitle = {Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies},
  year = {2021},
  doi = {10.18653/v1/2021.naacl-main.69},
  url = {https://doi.org/10.18653/v1/2021.naacl-main.69}
}

References

29 references, 6 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Yunmo Chen, Tongfei Chen, Seth Ebner, Aaron Steven White, and Benjamin Van Durme. 2020. Reading the manual: Event extraction as definition comprehension. In Proceedings of the Fourth Workshop on Structured Prediction for NLP, pages 74–83, Online. Association for Computational Linguistics. find
  2. Shumin Deng, Ningyu Zhang, Jiaojian Kang, Yichi Zhang, Wei Zhang, and Huajun Chen. 2020. Meta-learning with dynamic-memory-based prototypical network for few-shot event detection. WSDM. find
  3. Xinya Du and Claire Cardie. 2020. Event extraction by answering (almost) natural questions. In Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP), pages 671–683, Online. Association for Computational Linguistics. find our card
  4. Xinya Du, Alexander M. Rush, and Claire Cardie. 2021. Grit: Generative role-filler transformers for document-level event entity extraction. find
  5. Seth Ebner, Patrick Xia, Ryan Culkin, Kyle Rawlins, and Benjamin Van Durme. 2020. Multi-sentence argument linking. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pages 8057–8077, Online. Association for Computational Linguistics. find our card
  6. Rui Feng, Jie Yuan, and Chao Zhang. 2020. Probing and fine-tuning reading comprehension models for few-shot event extraction. In ArXiv. find
  7. Matthew Gerber and Joyce Chai. 2010. Beyond Nom-Bank: A study of implicit arguments for nominal predicates. In Proceedings of the 48th Annual Meeting of the Association for Computational Linguistics, pages 1583–1592, Uppsala, Sweden. Association for Computational Linguistics. find
  8. Ralph Grishman and Beth Sundheim. 1996. Message Understanding Conference- 6: A brief history. In COLING 1996 Volume 1: The 16th International Conference on Computational Linguistics. find
  9. Yutai Hou, Wanxiang Che, Yongkui Lai, Zhihan Zhou, Yijia Liu, Han Liu, and Ting Liu. 2020. Few-shot slot tagging with collapsed dependency transfer and label-enhanced task-adaptive projection network. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pages 1… find
  10. Lifu Huang, Heng Ji, Kyunghyun Cho, Ido Dagan, Sebastian Riedel, and Clare Voss. 2018. Zero-shot transfer learning for event extraction. In Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 2160–2170, Melbourne, Australia. Associat… find
  11. Heng Ji and Ralph Grishman. 2008. Refining event extraction through cross-document inference. In Proceedings of ACL-08: HLT, pages 254–262, Columbus, Ohio. Association for Computational Linguistics. find
  12. Carol C Kuhlthau. 1991. Inside the search process: Information seeking from the user’s perspective. Journal of the American society for information science, 42(5):361–371. find
  13. Viet Dac Lai, Thien Huu Nguyen, and Franck Dernoncourt. 2020. Extensively matching for few-shot learning event detection. In Proceedings of the First Joint Workshop on Narrative Understanding, Storylines, and Events, pages 38–45, Online. Association for Computational Linguistics. find
  14. Mike Lewis, Yinhan Liu, Naman Goyal, Marjan Ghazvininejad, Abdelrahman Mohamed, Omer Levy, Veselin Stoyanov, and Luke Zettlemoyer. 2020. BART: Denoising sequence-to-sequence pre-training for natural language generation, translation, and comprehension. In Proceedings of the 58th Annual Meeting of the… find our card
  15. Manling Li, Ying Lin, Tuan Manh Lai, Xiaoman Pan, Haoyang Wen, Sha Li, Zhenhailong Wang, Pengfei Yu, Lifu Huang, Di Lu, Qingyun Wang, Haoran Zhang, Qi Zeng, Chi Han, Zixuan Zhang, Yujia Qin, Xiaodan Hu, Nikolaus Parulian, Daniel Campos, Heng Ji, Brian Chen, Xudong Lin, Alireza Zareian, Amith Ananthr… find
  16. Ying Lin, Heng Ji, Fei Huang, and Lingfei Wu. 2020. A joint neural model for information extraction with global features. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pages 7999–8009, Online. Association for Computational Linguistics. find our card
  17. Jens-Erik Mai. 2016. Looking for information: A survey of research on information seeking, needs, and behavior. Emerald Group Publishing. find
  18. Yu Meng, Yunyi Zhang, Jiaxin Huang, Chenyan Xiong, Heng Ji, Chao Zhang, and Jiawei Han. 2020. Text classification using label names only: A language model self-training approach. In Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP), pages 9006–9017, Onlin… find
  19. Peng Shi and Jimmy Lin. 2019. Simple bert models for relation extraction and semantic role labeling. find our card
  20. Vered Shwartz, Peter West, Ronan Le Bras, Chandra Bhagavatula, and Yejin Choi. 2020. Unsupervised commonsense question answering with self-talk. In Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP), pages 4615–4629, Online. Association for Computational L… find
  21. Jake Snell, Kevin Swersky, and Richard Zemel. 2017. Prototypical networks for few-shot learning. NeurIPS. find
  22. David Wadden, Ulme Wennberg, Yi Luan, and Hannaneh Hajishirzi. 2019. Entity, relation, and event extraction with contextualized span representations. In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Langua… find our card
  23. Sung Whan Yoon, Jun Seo, and Jaekyun Moon. 2019. Tapnet: Neural network augmented with task-adaptive projection for few-shot learning. In ICML. find
  24. Zhisong Zhang, Xiang Kong, Zhengzhong Liu, Xuezhe Ma, and Eduard Hovy. 2020. A two-step approach for implicit event argument detection. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pages 7479–7485, Online. Association for Computational Linguistics. find
  25. Tatjana Moor, Michael Roth, and Anette Frank. 2013. Predicate-specific annotations for implicit role binding: Corpus annotation, data analysis and evaluation experiments. In Proceedings of the 10th International Conference on Computational Semantics (IWCS 2013) – Short Papers, pages 369–375, Potsdam… find
  26. Haoruo Peng, Yangqiu Song, and Dan Roth. 2016. Event detection and co-reference with minimal supervision. In Proceedings of the 2016 Conference on Empirical Methods in Natural Language Processing, pages 392–402, Austin, Texas. Association for Computational Linguistics. find
  27. Colin Raffel, Noam Shazeer, Adam Roberts, Katherine Lee, Sharan Narang, Michael Matena, Yanqi Zhou, W. Li, and Peter J. Liu. 2020. Exploring the limits of transfer learning with a unified text-to-text transformer. J. Mach. Learn. Res., 21:140:1–140:67. find
  28. Josef Ruppenhofer, Caroline Sporleder, Roser Morante, Collin Baker, and Martha Palmer. 2010. SemEval-2010 task 10: Linking events and their participants in discourse. In Proceedings of the 5th International Workshop on Semantic Evaluation, pages 45–50, Uppsala, Sweden. Association for Computational… find
  29. Document-Level Event Argument Extraction by Conditional Generation: Appendix Sha Li and Heng Ji and Jiawei Han University of Illinois at Urbana-Champaign, IL, USA {shal2, hengji, hanj}@illinois.edu Trigger Extraction Model Details 1.1 Tagging Scheme projection matrix M so that: M (c k ) = λM (φ k −… find