---
title: "BiCon-Gate — paper by Park and Zubiaga (2026)"
description: "They build a cautious rewrite in cumulative stages: contraction expansion, punctuation restoration, true-casing, and pronoun… 21 references."
authors: ["Hyunkyung Park", "Arkaitz Zubiaga"]
year: 2026
venue: "Proceedings of the Ninth Fact Extraction and VERification Workshop (FEVER)"
source: "https://aclanthology.org/2026.fever-1.5/"
doi: "10.18653/v1/2026.fever-1.5"
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/park-2026-bicon-gate-consistency-gated-de/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/park-2026-bicon-gate-consistency-gated-de/)

[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/25/) / BiCon-Gate: Consistency-Gated De-colloquialisation for Dialogue Fact-Checking

# BiCon-Gate: Consistency-Gated De-colloquialisation for Dialogue Fact-Checking

**Authors:** [Hyunkyung Park](https://md.research.elamur.ai/free/en/authors/hyunkyung-park/) and [Arkaitz Zubiaga](https://md.research.elamur.ai/free/en/authors/arkaitz-zubiaga/) · **Year:** 2026 · **Venue:** [Proceedings of the Ninth Fact Extraction and VERification Workshop (FEVER)](https://md.research.elamur.ai/free/en/venues/fever/) ([FEVER 2026](https://md.research.elamur.ai/free/en/venues/fever/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.fever-1.5/) · [PDF](https://aclanthology.org/2026.fever-1.5.pdf) · [DOI 10.18653/v1/2026.fever-1.5](https://doi.org/10.18653/v1/2026.fever-1.5) · [OpenAlex](https://openalex.org/W7140086918)

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

## Summary
Dialogue fact-checking breaks down when response claims are informal. Contractions, missing punctuation, lowercase text and pronouns that point to earlier turns make claims weak retrieval queries and shaky hypotheses for the verifier. Rewriting a claim so it stands on its own can raise recall, but it risks semantic drift (for example, a pronoun mapped to the wrong entity), which damages verification. Park and Zubiaga study this IR–FV trade-off on the DialFact benchmark.

They build a cautious rewrite in cumulative stages: contraction expansion, punctuation restoration, true-casing, and pronoun resolution restricted to the claim itself. They compare it with a one-shot rewrite by a decoder LLM. BiCon-Gate scores each rewrite using bidirectional NLI entailment and contradiction plus embedding similarity. It keeps the rewrite only above a task-specific threshold and otherwise reverts to the original claim.

Surface normalisation proves largely neutral, while scoped pronoun rewriting helps document retrieval. Gating yields the best FV-only results (especially on SUPPORTS) and the best top-1 E2E scores. One-shot LLM rewrites often hurt verification. The work shows that conservative, controlled rewriting is a safer component for retrieve–verify pipelines.

## Key points
- Across DialFact, 42.1% of claims include one or more in-claim pronouns (39.7% of validation claims, 44.3% of test claims), which motivates the focus on in-claim anaphora.
- The rewrite variants are cumulative: R1 adds apostrophes and expands contractions, R2 restores punctuation, R3 applies true-casing and R4 resolves in-scope pronouns. R5 is a single-pass de-colloquialisation by a decoder LLM.
- The gate score is α·ent + β·sim + γ·(1 − cnt). It uses thresholds of τ_IR = 0.50 for retrieval and τ_FV = 0.70 for verification, falling back to R0 below them.
- Effects are separated through IR-only, FV-only (gold evidence) and E2E evaluations. The E2E setting uses a fixed BM25 → E5 → BGE-CE retriever stack and a single top-1 Wikipedia passage as evidence.
- Ungated coreference rewrites can improve retrieval yet lower E2E verification, echoing earlier findings by Chamoun et al. (2023). The study is limited to English DialFact, and the thresholds may need retuning under distribution shift.

*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

- [University of London](https://md.research.elamur.ai/free/en/universities/university-of-london/) — University, London, United Kingdom · 13 works · 1,896 citations

📚 Shelves (6): [🖥️ Computer Science](https://md.research.elamur.ai/free/en/shelves/computer-science/) · 18,614 works · [🤖 Artificial Intelligence](https://md.research.elamur.ai/free/en/shelves/computer-science/artificial-intelligence/) · 17,965 works · [🎭 Arts and Humanities](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/) · 17,450 works · [🗣️ Language and Linguistics](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/language-and-linguistics/) · 17,296 works · [🤝 Social Sciences](https://md.research.elamur.ai/free/en/shelves/social-sciences/) · 1,954 works · [🤝 Sociology and Political Science](https://md.research.elamur.ai/free/en/shelves/social-sciences/sociology-and-political-science/) · 626 works

🧭 Topics (1): [🗨️ Dialogue systems](https://md.research.elamur.ai/free/en/topics/dialogue-systems/) · 654 works

## Cite

Cite as: Hyunkyung Park and Arkaitz Zubiaga. 2026. BiCon-Gate: Consistency-Gated De-colloquialisation for Dialogue Fact-Checking. Proceedings of the Ninth Fact Extraction and VERification Workshop (FEVER).

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

```bibtex
@inproceedings{park2026bicon,
  title = {{BiCon-Gate: Consistency-Gated De-colloquialisation for Dialogue Fact-Checking}},
  author = {Hyunkyung Park and Arkaitz Zubiaga},
  booktitle = {Proceedings of the Ninth Fact Extraction and VERification Workshop (FEVER)},
  year = {2026},
  doi = {10.18653/v1/2026.fever-1.5},
  url = {https://doi.org/10.18653/v1/2026.fever-1.5}
}
```

- [← previous Bibliography of Ethiopian Semitic, Cushitic, and Omotic Linguistics XXVIII: 2024](https://md.research.elamur.ai/free/en/papers/krzyzanowska-2026-bibliography-ethiopian-semitic-cushitic-omotic/)
- [next → Bidirectional Semantic Enhancement for Schema Routing Across Large-Scale Databases](https://md.research.elamur.ai/free/en/papers/wu-2026-bidirectional-semantic-enhancement-schema-routing/)
- [↑ back to papers of 2026](https://md.research.elamur.ai/free/en/papers/2026/page/25/)

## References

21 references, 9 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Xinnan Guo, Qian Zhu, Qiuhui Shi, Xuan Lin, Liubin Wang, DaqianLi DaqianLi, and Yongrui Chen. 2024. Context-aware tracking and dynamic introduction for incomplete utterance rewriting in extended multi-turn dialogues. In Findings of the Association for Computational Linguistics: ACL 2024, pages 2138–… [find](https://scholar.google.com/scholar?q=Xinnan+Guo%2C+Qian+Zhu%2C+Qiuhui+Shi%2C+Xuan+Lin%2C+Liubin+Wang%2C+DaqianLi+DaqianLi%2C+and+Yongrui+Chen%2E+2024%2E+Context%2Daware+tracking+and+dynamic+introduction+for+incomplete+utterance+rewriting+in+extended+multi)
2. Zhiyu Cao, Peifeng Li, Yaxin Fan, and Qiaoming Zhu. 2024. Incomplete utterance rewriting with editing operation guidance and utterance augmentation. In Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing, pages 7225– 7238, Miami, Florida, USA. Association for Compu… [find](https://scholar.google.com/scholar?q=Zhiyu+Cao%2C+Peifeng+Li%2C+Yaxin+Fan%2C+and+Qiaoming+Zhu%2E+2024%2E+Incomplete+utterance+rewriting+with+editing+operation+guidance+and+utterance+augmentation%2E+In+Proceedings+of+the+2024+Conference+on+Empirical+)
3. Prakhar Gupta, Chien-Sheng Wu, Wenhao Liu, and Caiming Xiong. 2022. [DialFact: A benchmark for fact-checking in dialogue](https://md.research.elamur.ai/free/en/papers/gupta-2022-dialfact-benchmark-fact-checking-dialogue/). In Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 3785–3801, Dublin, Ireland. Association for Computational…
4. In Proceedings of the 31st International Conference on Computational Linguistics, pages 9721–9736. Qwen, An Yang, Baosong Yang, Beichen Zhang, Binyuan Hui, Bo Zheng, Bowen Yu, Chengyuan Li, Dayiheng Liu, Fei Huang, Haoran Wei, Huan Lin, Jian Yang, Jianhong Tu, Jianwei Zhang, Jianxin Yang, Jiaxi Yang… [find](https://scholar.google.com/scholar?q=In+Proceedings+of+the+31st+International+Conference+on+Computational+Linguistics%2C+pages+9721%E2%80%939736%2E+Qwen%2C+An+Yang%2C+Baosong+Yang%2C+Beichen+Zhang%2C+Binyuan+Hui%2C+Bo+Zheng%2C+Bowen+Yu%2C+Chengyuan+Li%2C+Dayiheng+L)
5. Pengcheng He, Jianfeng Gao, and Weizhu Chen. 2023. [Debertav3: Improving deberta using electra-style pre-training with gradient-disentangled embedding sharing](https://md.research.elamur.ai/free/en/papers/he-2021-debertav3-improving-deberta-using-electra/). Preprint, arXiv:2111.09543.
6. Megha Sundriyal, Tanmoy Chakraborty, and Preslav Nakov. 2023. From chaos to clarity: Claim normalization to empower fact-checking. In Findings of the Association for Computational Linguistics: EMNLP 2023, pages 6594–6609, Singapore. Association for Computational Linguistics. [find](https://scholar.google.com/scholar?q=Megha+Sundriyal%2C+Tanmoy+Chakraborty%2C+and+Preslav+Nakov%2E+2023%2E+From+chaos+to+clarity%3A+Claim+normalization+to+empower+fact%2Dchecking%2E+In+Findings+of+the+Association+for+Computational+Linguistics%3A+EMNLP+2)
7. Byeongchang Kim, Hyunwoo Kim, Seokhee Hong, and Gunhee Kim. 2021. How robust are fact checking systems on colloquial claims? In Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, pages 1535–1548, Online. Ass… [find](https://scholar.google.com/scholar?q=Byeongchang+Kim%2C+Hyunwoo+Kim%2C+Seokhee+Hong%2C+and+Gunhee+Kim%2E+2021%2E+How+robust+are+fact+checking+systems+on+colloquial+claims%3F+In+Proceedings+of+the+2021+Conference+of+the+North+American+Chapter+of+the+)
8. James Thorne and Andreas Vlachos. 2018. [Automated fact checking: Task formulations, methods and future directions](https://md.research.elamur.ai/free/en/papers/thorne-2018-automated-fact-checking-task-formulations/). In Proceedings of the 27th International Conference on Computational Linguistics, pages 3346–3359, Santa Fe, New Mexico, USA. Association for Computational Linguistics.
9. Philippe Laban, Tobias Schnabel, Paul N Bennett, and Marti A Hearst. 2022. Summac: Re-visiting nli-based models for inconsistency detection in summarization. Transactions of the Association for Computational Linguistics, 10:163–177. [find](https://scholar.google.com/scholar?q=Philippe+Laban%2C+Tobias+Schnabel%2C+Paul+N+Bennett%2C+and+Marti+A+Hearst%2E+2022%2E+Summac%3A+Re%2Dvisiting+nli%2Dbased+models+for+inconsistency+detection+in+summarization%2E+Transactions+of+the+Association+for+Comput)
10. James Thorne, Andreas Vlachos, Christos Christodoulopoulos, and Arpit Mittal. 2018. [FEVER: a large-scale dataset for fact extraction and VERification](https://md.research.elamur.ai/free/en/papers/thorne-2018-fever-large-scale-dataset-fact/). In Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Vo…
11. Moritz Laurer, Wouter van Atteveldt, Andreu Casas, and Kasper Welbers. 2024. Building efficient universal classifiers with natural language inference. Preprint, arXiv:2312.17543. — [arXiv:2312.17543](https://arxiv.org/abs/2312.17543) [find](https://scholar.google.com/scholar?q=Moritz+Laurer%2C+Wouter+van+Atteveldt%2C+Andreu+Casas%2C+and+Kasper+Welbers%2E+2024%2E+Building+efficient+universal+classifiers+with+natural+language+inference%2E+Preprint%2C+arXiv%3A2312%2E17543%2E+%E2%80%94+%5BarXiv%3A2312%2E17543%5D%28)
12. Zitong Li, Jiawei Li, Haifeng Tang, Kenny Zhu, and Ruolan Yang. 2023. Incomplete utterance rewriting by a two-phase locate-and-fill regime. In Findings of the Association for Computational Linguistics: ACL 2023, pages 2731–2745, Toronto, Canada. Association for Computational Linguistics. [find](https://scholar.google.com/scholar?q=Zitong+Li%2C+Jiawei+Li%2C+Haifeng+Tang%2C+Kenny+Zhu%2C+and+Ruolan+Yang%2E+2023%2E+Incomplete+utterance+rewriting+by+a+two%2Dphase+locate%2Dand%2Dfill+regime%2E+In+Findings+of+the+Association+for+Computational+Linguistics)
13. Vincent Vandeghinste and Oliver Guhr. 2024. Fullstop: Punctuation and segmentation prediction for dutch with transformers. Language Resources and Evaluation, 58(4):1335–1354. [find](https://scholar.google.com/scholar?q=Vincent+Vandeghinste+and+Oliver+Guhr%2E+2024%2E+Fullstop%3A+Punctuation+and+segmentation+prediction+for+dutch+with+transformers%2E+Language+Resources+and+Evaluation%2C+58%284%29%3A1335%E2%80%931354%2E)
14. Jimmy Lin, Xueguang Ma, Sheng-Chieh Lin, Jheng-Hong Yang, Ronak Pradeep, and Rodrigo Nogueira. 2021. Pyserini: An easy-to-use python toolkit to support replicable ir research with sparse and dense representations. Preprint, arXiv:2102.10073. — [arXiv:2102.10073](https://arxiv.org/abs/2102.10073) [find](https://scholar.google.com/scholar?q=Jimmy+Lin%2C+Xueguang+Ma%2C+Sheng%2DChieh+Lin%2C+Jheng%2DHong+Yang%2C+Ronak+Pradeep%2C+and+Rodrigo+Nogueira%2E+2021%2E+Pyserini%3A+An+easy%2Dto%2Duse+python+toolkit+to+support+replicable+ir+research+with+sparse+and+dense+rep)
15. Liang Wang, Nan Yang, Xiaolong Huang, Linjun Yang, Rangan Majumder, and Furu Wei. 2024. [Multilingual e5 text embeddings: A technical report](https://md.research.elamur.ai/free/en/papers/wang-2024-multilingual-e5-text-embeddings-technical/). Preprint, arXiv:2402.05672.
16. Johnathan Xie, Annie Chen, Yoonho Lee, Eric Mitchell, and Chelsea Finn. 2024. Calibrating language models with adaptive temperature scaling. In Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing, pages 18128–18138. [find](https://scholar.google.com/scholar?q=Johnathan+Xie%2C+Annie+Chen%2C+Yoonho+Lee%2C+Eric+Mitchell%2C+and+Chelsea+Finn%2E+2024%2E+Calibrating+language+models+with+adaptive+temperature+scaling%2E+In+Proceedings+of+the+2024+Conference+on+Empirical+Methods+)
17. Kawshik Manikantan, Makarand Tapaswi, Vineet Gandhi, and Shubham Toshniwal. 2025. [IdentifyMe: A challenging long-context mention resolution benchmark for LLMs](https://md.research.elamur.ai/free/en/papers/manikantan-2025-identifyme-challenging-long-context-mention/). In Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Languag…
18. Yuheng Zha, Yichi Yang, Ruichen Li, and Zhiting Hu. 2023. [Alignscore: Evaluating factual consistency with a unified alignment function](https://md.research.elamur.ai/free/en/papers/zha-2023-alignscore-evaluating-factual-consistency-unified/). In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 11328–11348.
19. Giuliano Martinelli, Edoardo Barba, and Roberto Navigli. 2024. [Maverick: Efficient and accurate coreference resolution defying recent trends](https://md.research.elamur.ai/free/en/papers/martinelli-2024-maverick-efficient-accurate-coreference-resolution/). In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 13380–13394, Bangkok, Thailand. Asso…
20. Hengran Zhang, Ruqing Zhang, Jiafeng Guo, Maarten de Rijke, Yixing Fan, and Xueqi Cheng. 2023. From relevance to utility: Evidence retrieval with feedback for fact verification. In Findings of the Association for Computational Linguistics: EMNLP 2023, pages 6373–6384, Singapore. Association for Comp… [find](https://scholar.google.com/scholar?q=Hengran+Zhang%2C+Ruqing+Zhang%2C+Jiafeng+Guo%2C+Maarten+de+Rijke%2C+Yixing+Fan%2C+and+Xueqi+Cheng%2E+2023%2E+From+relevance+to+utility%3A+Evidence+retrieval+with+feedback+for+fact+verification%2E+In+Findings+of+the+Ass)
21. Dina Pisarevskaya and Arkaitz Zubiaga. 2025. [Zero-shot and few-shot learning with instruction-following llms for claim matching in automated fact-checking](https://md.research.elamur.ai/free/en/papers/pisarevskaya-2025-zero-shot-few-shot-learning/).

*Updated 2 November 2026*