---
title: "Constructing a Silver Corpus for Weakly Supervised Vietnamese"
description: "Vietnamese event extraction (EE) has little annotated data. 12 references. BKEE, the main benchmark, is small and not very diverse."
authors: ["Phạm Xuân Hiệu", "Tuan Vu Minh", "Mai-Vu Tran", "Hoang-Quynh Le"]
year: 2026
venue: "Proceedings of the 9th Workshop on Event Extraction and Understanding: Challenges and Applications (EEUCA 2026)"
source: "https://aclanthology.org/2026.eeuca-1.4/"
doi: "10.18653/v1/2026.eeuca-1.4"
licence: "CC BY 4.0 (ACL Anthology)"
licence_url: "https://creativecommons.org/licenses/by/4.0/"
summary_licence: "Elamur Cosmic License"
canonical: "https://research.elamur.ai/free/en/papers/hieu-2026-constructing-silver-corpus-weakly-supervised/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/hieu-2026-constructing-silver-corpus-weakly-supervised/)

[home](https://md.research.elamur.ai/free/en/) / [papers](https://md.research.elamur.ai/free/en/papers/) / [2026](https://md.research.elamur.ai/free/en/papers/2026/page/26/) / Constructing a Silver Corpus for Weakly Supervised Vietnamese Event Extraction using Cross-Document N-ary Relation Filtering

# Constructing a Silver Corpus for Weakly Supervised Vietnamese Event Extraction using Cross-Document N-ary Relation Filtering

**Authors:** Phạm Xuân Hiệu, Tuan Vu Minh, Mai-Vu Tran and [Hoang-Quynh Le](https://md.research.elamur.ai/free/en/authors/hoang-quynh-le/) · **Year:** 2026 · **Venue:** [Proceedings of the 9th Workshop on Event Extraction and Understanding: Challenges and Applications (EEUCA 2026)](https://md.research.elamur.ai/free/en/venues/eeuca/) ([Workshop on Event Extraction and Understanding: Challenges and Applications 2026](https://md.research.elamur.ai/free/en/venues/eeuca/2026/)) · **Publisher:** [Association for Computational Linguistics](https://md.research.elamur.ai/free/en/publishers/association-for-computational-linguistics/) · **Licence:** [CC BY 4.0 (ACL Anthology)](https://creativecommons.org/licenses/by/4.0/)

**Links:** [Publisher's page](https://aclanthology.org/2026.eeuca-1.4/) · [PDF](https://aclanthology.org/2026.eeuca-1.4.pdf) · [DOI 10.18653/v1/2026.eeuca-1.4](https://doi.org/10.18653/v1/2026.eeuca-1.4) · [OpenAlex](https://openalex.org/W7166833677)

**Citations:** 0 citations (OpenAlex, 2026-10-14)

## Summary

Vietnamese event extraction (EE) has little annotated data. BKEE, the main benchmark, is small and not very diverse. The authors address this by building a large silver-standard corpus without manual annotation.

They collected about 72,000 Vietnamese news articles from the aggregators Báo Mới and Google News, keeping the platforms' topic groupings. The existing BKEE model then pseudo-labeled every sentence. To remove noisy labels, they use the fact that one real-world event is usually covered by several outlets. They extract n-ary relations (a trigger plus its arguments) and keep only those that recur across documents in the same topic group. Next, a schema-based augmentation step, adapted to single sentences, recombines and extends event schemas. GPT-4o then turns these schemas into new, structurally valid sentences.

The resulting corpus was used to train the joint FourIE model. On BKEE, entity mention detection and argument extraction improved, while event detection stayed roughly level. The work shows that careful weak supervision can partly replace costly annotation for low-resource EE.

## Key points

- **Raw data:** more than 300 topic groups containing 2,673,796 unlabeled sentences, pseudo-labeled with the BKEE model.
- **Filtering method:** works at two frequency levels.
  - Event types must appear in at least µ sentences of a group.
  - Individual relations must pass an adaptive per-type threshold based on IQR (λ = 3).
  - After filtering, 15,260 sentences remain.
- **Final corpus:** after augmentation, the silver corpus has 46,240 sentences. It is released as VietEE2 at github.com/Larken1612/VietEE2.
- **Results:** with FourIE as the backbone, F1 rose by +3.3 on EMD and +2.2 on EAE over strong baselines, and ED stayed competitive. Ablations show that removing either the filtering or the augmentation hurts performance.
- **Training strategy:** a curriculum-style approach worked best. The model is first pre-trained on BKEE plus 20% of the silver data, then trained on the rest. This beat both training from scratch on all data and pre-training on gold data alone.

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

## Institutions

- [VNU University of Engineering and Technology](https://md.research.elamur.ai/free/en/universities/vnu-university-of-engineering-and-technology/) — University, Hanoi, Vietnam · 4 works · 1 citation

📚 Shelves (4): [🖥️ Computer Science](https://md.research.elamur.ai/free/en/shelves/computer-science/) · 16,667 works · [🤖 Artificial Intelligence](https://md.research.elamur.ai/free/en/shelves/computer-science/artificial-intelligence/) · 16,194 works · [🎭 Arts and Humanities](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/) · 16,039 works · [🗣️ Language and Linguistics](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/language-and-linguistics/) · 15,891 works

🧭 Topics (1): [🔎 Information extraction](https://md.research.elamur.ai/free/en/topics/information-extraction/) · 167 works

## Cite

Cite as: Phạm Xuân Hiệu, Tuan Vu Minh, Mai-Vu Tran and Hoang-Quynh Le. 2026. Constructing a Silver Corpus for Weakly Supervised Vietnamese Event Extraction using Cross-Document N-ary Relation Filtering. Proceedings of the 9th Workshop on Event Extraction and Understanding: Challenges and Applications (EEUCA 2026).

The paper itself, not this page. BibTeX from the library's record of the work.

```bibtex
@inproceedings{hieu2026constructing,
  title = {{Constructing a Silver Corpus for Weakly Supervised Vietnamese Event Extraction using Cross-Document N-ary Relation Filtering}},
  author = {Phạm Xuân Hiệu, Tuan Vu Minh, Mai-Vu Tran and Hoang-Quynh Le},
  booktitle = {Proceedings of the 9th Workshop on Event Extraction and Understanding: Challenges and Applications (EEUCA 2026)},
  year = {2026},
  doi = {10.18653/v1/2026.eeuca-1.4},
  url = {https://doi.org/10.18653/v1/2026.eeuca-1.4}
}
```

- [← previous Constructing a Japanese Verdict Prediction Dataset for Fact-Checking of LLM-Generated Texts](https://md.research.elamur.ai/free/en/papers/masano-2026-constructing-japanese-verdict-prediction-dataset/)
- [next → Construction and Analysis of Japanese Parent-Child Dialogic Reading Corpus for Conversational Agents](https://md.research.elamur.ai/free/en/papers/nakagi-2026-construction-analysis-japanese-parent-child/)
- [↑ back to papers of 2026](https://md.research.elamur.ai/free/en/papers/2026/page/26/)

## References

12 references, 4 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Jun Araki and Teruko Mitamura. 2018. Open-domain event detection using distant supervision. In Proceedings of the 27th International Conference on Computational Linguistics, pages 878–891. Association for Computational Linguistics. [find](https://scholar.google.com/scholar?q=Jun+Araki+and+Teruko+Mitamura%2E+2018%2E+Open%2Ddomain+event+detection+using+distant+supervision%2E+In+Proceedings+of+the+27th+International+Conference+on+Computational+Linguistics%2C+pages+878%E2%80%93891%2E+Association)
2. Ying Lin, Heng Ji, Fei Huang, and Lingfei Wu. 2020. [A joint neural model for information extraction with global features](https://md.research.elamur.ai/free/en/papers/lin-2020-joint-neural-model-information-extraction/). In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pages 7999–8009. Association for Computational Linguistics.
3. Alexis Conneau, Kartikay Khandelwal, Naman Goyal, Vishrav Chaudhary, Guillaume Wenzek, Francisco Guzmán, Edouard Grave, Myle Ott, Luke Zettlemoyer, and Veselin Stoyanov. 2019. Unsupervised cross-lingual representation learning at scale. CoRR, abs/1911.02116. [find](https://scholar.google.com/scholar?q=Alexis+Conneau%2C+Kartikay+Khandelwal%2C+Naman+Goyal%2C+Vishrav+Chaudhary%2C+Guillaume+Wenzek%2C+Francisco+Guzm%C3%A1n%2C+Edouard+Grave%2C+Myle+Ott%2C+Luke+Zettlemoyer%2C+and+Veselin+Stoyanov%2E+2019%2E+Unsupervised+cross%2Dlingu)
4. Kang Liu, Yubo Chen, Jian Liu, Xinyu Zuo, and Jun Zhao. 2020. Extracting events and their relations from texts: A survey on recent research progress and challenges. AI Open, 1:22–39. [find](https://scholar.google.com/scholar?q=Kang+Liu%2C+Yubo+Chen%2C+Jian+Liu%2C+Xinyu+Zuo%2C+and+Jun+Zhao%2E+2020%2E+Extracting+events+and+their+relations+from+texts%3A+A+survey+on+recent+research+progress+and+challenges%2E+AI+Open%2C+1%3A22%E2%80%9339%2E)
5. Teruko Mitamura, Zhengzhong Liu, and Eduard H Hovy. 2017. Events detection, coreference and sequencing: What’s next? overview of the tac kbp 2017 event track. In TAC. [find](https://scholar.google.com/scholar?q=Teruko+Mitamura%2C+Zhengzhong+Liu%2C+and+Eduard+H+Hovy%2E+2017%2E+Events+detection%2C+coreference+and+sequencing%3A+What%E2%80%99s+next%3F+overview+of+the+tac+kbp+2017+event+track%2E+In+TAC%2E)
6. Xiang Dai and Heike Adel. 2020. [An analysis of simple data augmentation for named entity recognition](https://md.research.elamur.ai/free/en/papers/dai-2020-analysis-simple-data-augmentation-named/). In Proceedings of the 28th International Conference on Computational Linguistics, pages 3861–3867. International Committee on Computational Linguistics.
7. Dat Quoc Nguyen and Anh-Tuan Nguyen. 2020. [Phobert: Pre-trained language models for vietnamese](https://md.research.elamur.ai/free/en/papers/nguyen-2020-phobert-pre-trained-language-models/). In Findings of the association for computational linguistics: EMNLP 2020, pages 1037–1042.
8. Seth Ebner, Patrick Xia, Ryan Culkin, Kyle Rawlins, and Benjamin Van Durme. 2020. [Multi-sentence argument linking](https://md.research.elamur.ai/free/en/papers/ebner-2020-multi-sentence-argument-linking/). In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pages 8057–8077. Association for Computational Linguistics.
9. Minh Van Nguyen, Viet Dac Lai, and Thien Huu Nguyen. 2021. Cross-task instance representation interactions and label dependencies for joint information extraction with graph convolutional networks. [find](https://scholar.google.com/scholar?q=Minh+Van+Nguyen%2C+Viet+Dac+Lai%2C+and+Thien+Huu+Nguyen%2E+2021%2E+Cross%2Dtask+instance+representation+interactions+and+label+dependencies+for+joint+information+extraction+with+graph+convolutional+networks%2E)
10. In Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, pages 27–38. Association for Computational Linguistics. [find](https://scholar.google.com/scholar?q=In+Proceedings+of+the+2021+Conference+of+the+North+American+Chapter+of+the+Association+for+Computational+Linguistics%3A+Human+Language+Technologies%2C+pages+27%E2%80%9338%2E+Association+for+Computational+Linguistic)
11. Thi-Nhung Nguyen, Bang Tien Tran, Trong-Nghia Luu, Thien Huu Nguyen, and Kiem-Hieu Nguyen. 2024. BKEE: Pioneering event extraction in the Vietnamese language. In Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024),… [find](https://scholar.google.com/scholar?q=Thi%2DNhung+Nguyen%2C+Bang+Tien+Tran%2C+Trong%2DNghia+Luu%2C+Thien+Huu+Nguyen%2C+and+Kiem%2DHieu+Nguyen%2E+2024%2E+BKEE%3A+Pioneering+event+extraction+in+the+Vietnamese+language%2E+In+Proceedings+of+the+2024+Joint+Internat)
12. Thanh Vu, Dat Quoc Nguyen, Dai Quoc Nguyen, Mark Dras, and Mark Johnson. 2018. VnCoreNLP: A Vietnamese natural language processing toolkit. In Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Demonstrations, pages 56–60. Association f… [find](https://scholar.google.com/scholar?q=Thanh+Vu%2C+Dat+Quoc+Nguyen%2C+Dai+Quoc+Nguyen%2C+Mark+Dras%2C+and+Mark+Johnson%2E+2018%2E+VnCoreNLP%3A+A+Vietnamese+natural+language+processing+toolkit%2E+In+Proceedings+of+the+2018+Conference+of+the+North+American+)

*Updated 1 November 2026*