---
title: "Neural Fuzzy Repair — paper by Bulte and Tezcan (2019)"
description: "Bulté and Tezcan propose Neural Fuzzy Repair (NFR), which brings TM knowledge into NMT purely… 34 references, cited by 55 papers here."
authors: ["Bram Bulte", "Arda Tezcan"]
year: 2019
venue: "Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics"
source: "https://aclanthology.org/P19-1175/"
doi: "10.18653/v1/p19-1175"
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/bulte-2019-neural-fuzzy-repair-integrating-fuzzy/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/bulte-2019-neural-fuzzy-repair-integrating-fuzzy/)

[home](https://md.research.elamur.ai/free/en/) / [papers](https://md.research.elamur.ai/free/en/papers/) / [2019](https://md.research.elamur.ai/free/en/papers/2019/page/5/) / Neural Fuzzy Repair: Integrating Fuzzy Matches into Neural Machine Translation

# Neural Fuzzy Repair: Integrating Fuzzy Matches into Neural Machine Translation

**Authors:** Bram Bulte and [Arda Tezcan](https://md.research.elamur.ai/free/en/authors/arda-tezcan/) · **Year:** 2019 · **Venue:** [Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics](https://md.research.elamur.ai/free/en/venues/acl/) ([ACL 2019](https://md.research.elamur.ai/free/en/venues/acl/2019/)) · **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/P19-1175/) · [PDF](https://aclanthology.org/P19-1175.pdf) · [DOI 10.18653/v1/p19-1175](https://doi.org/10.18653/v1/p19-1175) · [OpenAlex](https://openalex.org/W2952317054)

**Citations:** 65 citations (OpenAlex, 2026-10-14) · [55](https://md.research.elamur.ai/free/en/papers/bulte-2019-neural-fuzzy-repair-integrating-fuzzy/cited-by/) in our library, of the 141,840 papers whose bibliographies we read

## Summary
Translation memories (TMs) stay valuable in professional translation because close matches from earlier work often need little editing. In typical CAT workflows, however, NMT only serves as a fallback when the TM offers no good match. Bulté and Tezcan propose Neural Fuzzy Repair (NFR), which brings TM knowledge into NMT purely through data preprocessing. For each sentence, the TM is searched for comparable source-side entries using token-level edit distance, sped up by SetSimilaritySearch candidate filtering. The target sides of the best matches are then appended to the source, separated by a break token. A standard OpenNMT model is trained on this data in one of two ways: as a dedicated system paired with a backoff NMT model, or as a single unified system using both original and augmented data. Experiments on DGT-TM for EN→NL and EN→HU show consistent, sizeable gains over NMT, Moses SMT, plain TM matching, a hybrid TM-SMT system and Google Translate. Because the architecture is unchanged, the method fits any NMT toolkit wherever a large, repetitive TM exists.

## Key points
- **Data:** DGT-TM (European Commission) for English→Dutch and English→Hungarian.
  - About 2.4M training pairs, two dev sets of 3000 pairs each, and a 3207-sentence test set.
  - Test sentences with exact matches were removed.
- **Augmentation formats:** the source is joined with the targets of the top 1, 2 or 3 fuzzy matches via "@@@". A "format 1 n-best" variant adds a separate training pair for each of the n best matches.
- **Model:** bidirectional RNN seq2seq with global attention in OpenNMT.
  - 750 LSTM units, 200K training steps, and a source vocabulary of 100K, since the input is bilingual.
- **Integration results:** dedicated NFR combined with backoff NMT scored best. Unified NFR reached comparable BLEU gains over baseline NMT.
- **When NFR helps:**
  - It starts to pay off at fuzzy match scores of roughly 0.5–0.6.
  - An NFR system trained on only 1/8 of the data still beat baseline NMT trained on the full set.
  - Baseline comparison differed by language: for EN-NL, SMT beat NMT by about 3 BLEU, while for EN-HU, NMT led SMT by about 7 BLEU.

*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

- [Ghent University](https://md.research.elamur.ai/free/en/universities/ghent-university/) — University, Ghent, Belgium · 18 works · 2,888 citations
- [KU Leuven](https://md.research.elamur.ai/free/en/universities/ku-leuven/) — University, Leuven, Belgium · 25 works · 5,716 citations

📚 Shelves (4): [🖥️ Computer Science](https://md.research.elamur.ai/free/en/shelves/computer-science/) · 9,085 works · [🤖 Artificial Intelligence](https://md.research.elamur.ai/free/en/shelves/computer-science/artificial-intelligence/) · 8,888 works · [🎭 Arts and Humanities](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/) · 9,296 works · [🗣️ Language and Linguistics](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/language-and-linguistics/) · 9,206 works

🧭 Topics (1): [🔁 Machine translation](https://md.research.elamur.ai/free/en/topics/machine-translation/) · 958 works

## Cite

Cite as: Bram Bulte and Arda Tezcan. 2019. Neural Fuzzy Repair: Integrating Fuzzy Matches into Neural Machine Translation. Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics.

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

```bibtex
@inproceedings{bulte2019neural,
  title = {{Neural Fuzzy Repair: Integrating Fuzzy Matches into Neural Machine Translation}},
  author = {Bram Bulte and Arda Tezcan},
  booktitle = {Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics},
  year = {2019},
  doi = {10.18653/v1/p19-1175},
  url = {https://doi.org/10.18653/v1/p19-1175}
}
```

- [← previous Collaborative Dialogue in Minecraft](https://md.research.elamur.ai/free/en/papers/chen-2019-collaborative-dialogue-minecraft/)
- [next → On Measuring Gender Bias in Translation of Gender-neutral Pronouns](https://md.research.elamur.ai/free/en/papers/cho-2019-measuring-gender-bias-translation-gender/)
- [↑ back to papers of 2019](https://md.research.elamur.ai/free/en/papers/2019/page/5/)

## References

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

1. Roberto J. Bayardo, Yiming Ma, and Ramakrishnan Srikant. 2007. Scaling up all pairs similarity search. In Proceedings of the 16th International Conference on World Wide Web, pages 131–140. [find](https://scholar.google.com/scholar?q=Roberto+J%2E+Bayardo%2C+Yiming+Ma%2C+and+Ramakrishnan+Srikant%2E+2007%2E+Scaling+up+all+pairs+similarity+search%2E+In+Proceedings+of+the+16th+International+Conference+on+World+Wide+Web%2C+pages+131%E2%80%93140%2E)
2. Ergun Biçici and Marc Dymetman. 2008. Dynamic translation memory: using statistical machine translation to improve translation memory fuzzy matches. In International Conference on Intelligent Text Processing and Computational Linguistics, pages 454– 465. [find](https://scholar.google.com/scholar?q=Ergun+Bic%CC%A7ici+and+Marc+Dymetman%2E+2008%2E+Dynamic+translation+memory%3A+using+statistical+machine+translation+to+improve+translation+memory+fuzzy+matches%2E+In+International+Conference+on+Intelligent+Text+Pr)
3. Michael Bloodgood and Benjamin Strauss. 2015. Translation memory retrieval methods. Computing Research Repository, arXiv:1505.05841. [find](https://scholar.google.com/scholar?q=Michael+Bloodgood+and+Benjamin+Strauss%2E+2015%2E+Translation+memory+retrieval+methods%2E+Computing+Research+Repository%2C+arXiv%3A1505%2E05841%2E)
4. Bram Bulté, Tom Vanallemeersch, and Vincent Vandeghinste. 2018. M3TRA: integrating TM and MT for professional translators. In Proceedings of the 21st Annual Conference of the European Association for Machine Translation, pages 69–78. [find](https://scholar.google.com/scholar?q=Bram+Bulte%CC%81%2C+Tom+Vanallemeersch%2C+and+Vincent+Vandeghinste%2E+2018%2E+M3TRA%3A+integrating+TM+and+MT+for+professional+translators%2E+In+Proceedings+of+the+21st+Annual+Conference+of+the+European+Association+for)
5. Patrick Cadwell, Sharon O’Brien, and Carlos S. C. Teixeira. 2018. Resistance and accommodation: factors for the (non-) adoption of machine translation among professional translators. Perspectives, 26(3):301–321. [find](https://scholar.google.com/scholar?q=Patrick+Cadwell%2C+Sharon+O%E2%80%99Brien%2C+and+Carlos+S%2E+C%2E+Teixeira%2E+2018%2E+Resistance+and+accommodation%3A+factors+for+the+%28non%2D%29+adoption+of+machine+translation+among+professional+translators%2E+Perspectives%2C+26%28)
6. Qian Cao and Deyi Xiong. 2018. Encoding gated translation memory into neural machine translation. In Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing, pages 3042–3047. [find](https://scholar.google.com/scholar?q=Qian+Cao+and+Deyi+Xiong%2E+2018%2E+Encoding+gated+translation+memory+into+neural+machine+translation%2E+In+Proceedings+of+the+2018+Conference+on+Empirical+Methods+in+Natural+Language+Processing%2C+pages+3042%E2%80%93)
7. Raj Dabre, Fabien Cromieres, and Sadao Kurohashi. 2017. Enabling multi-source neural machine translation by concatenating source sentences in multiple languages. Computing Research Repository, arXiv:1702.06135. [find](https://scholar.google.com/scholar?q=Raj+Dabre%2C+Fabien+Cromieres%2C+and+Sadao+Kurohashi%2E+2017%2E+Enabling+multi%2Dsource+neural+machine+translation+by+concatenating+source+sentences+in+multiple+languages%2E+Computing+Research+Repository%2C+arXiv%3A1)
8. Marcello Federico, Alessandro Cattelan, and Marco Trombetti. 2012. Measuring user productivity in machine translation enhanced Computer Assisted Translation. In Proceedings of the 2012 Conference of the Association for Machine Translation in the Americas, pages 44–56. [find](https://scholar.google.com/scholar?q=Marcello+Federico%2C+Alessandro+Cattelan%2C+and+Marco+Trombetti%2E+2012%2E+Measuring+user+productivity+in+machine+translation+enhanced+Computer+Assisted+Translation%2E+In+Proceedings+of+the+2012+Conference+of+t)
9. Yang Feng, Shiyue Zhang, Andi Zhang, Dong Wang, and Andrew Abel. 2017. Memory-augmented neural machine translation. Computing Research Repository, arXiv:1708.02005. [find](https://scholar.google.com/scholar?q=Yang+Feng%2C+Shiyue+Zhang%2C+Andi+Zhang%2C+Dong+Wang%2C+and+Andrew+Abel%2E+2017%2E+Memory%2Daugmented+neural+machine+translation%2E+Computing+Research+Repository%2C+arXiv%3A1708%2E02005%2E)
10. Jiatao Gu, Yong Wang, Kyunghyun Cho, and Victor O. K. Li. 2018. Search engine guided neural machine translation. In Proceedings of the 32nd AAAI Conference on Artificial Intelligence, pages 5133– 5140. [find](https://scholar.google.com/scholar?q=Jiatao+Gu%2C+Yong+Wang%2C+Kyunghyun+Cho%2C+and+Victor+O%2E+K%2E+Li%2E+2018%2E+Search+engine+guided+neural+machine+translation%2E+In+Proceedings+of+the+32nd+AAAI+Conference+on+Artificial+Intelligence%2C+pages+5133%E2%80%93+5140)
11. Sanjika Hewavitharana, Stephan Vogel, and Alex Waibel. 2005. Augmenting a statistical translation system with a translation memory. In Proceedings of the 10th Annual Conference of the European Association for Machine Translation, pages 126–132. [find](https://scholar.google.com/scholar?q=Sanjika+Hewavitharana%2C+Stephan+Vogel%2C+and+Alex+Waibel%2E+2005%2E+Augmenting+a+statistical+translation+system+with+a+translation+memory%2E+In+Proceedings+of+the+10th+Annual+Conference+of+the+European+Associa)
12. Chris Hokamp. 2017. Ensembling factored neural machine translation models for automatic post-editing and quality estimation. Computing Research Repository, arXiv:1706.05083. [find](https://scholar.google.com/scholar?q=Chris+Hokamp%2E+2017%2E+Ensembling+factored+neural+machine+translation+models+for+automatic+post%2Dediting+and+quality+estimation%2E+Computing+Research+Repository%2C+arXiv%3A1706%2E05083%2E)
13. Chris Hokamp and Qun Liu. 2017. [Lexically constrained decoding for sequence generation using grid beam search](https://md.research.elamur.ai/free/en/papers/hokamp-2017-lexically-constrained-decoding-sequence-generation/). Computing Research Repository, arXiv:1704.07138.
14. Heikki Hyyrö. 2001. Explaining and extending the bit-parallel approximate string matching algorithm of Myers. Technical report, Dept. of Computer and Information Sciences, University of Tampere. [find](https://scholar.google.com/scholar?q=Heikki+Hyyro%CC%88%2E+2001%2E+Explaining+and+extending+the+bit%2Dparallel+approximate+string+matching+algorithm+of+Myers%2E+Technical+report%2C+Dept%2E+of+Computer+and+Information+Sciences%2C+University+of+Tampere%2E)
15. Guillaume Klein, Yoon Kim, Yuntian Deng, Jean Senellart, and Alexander M Rush. 2017. [OpenNMT: Open-source toolkit for neural machine translation](https://md.research.elamur.ai/free/en/papers/klein-2017-opennmt-open-source-toolkit-neural/). Computing Research Repository, arXiv:1701.02810.
16. Philipp Koehn. 2004. [Statistical significance tests for machine translation evaluation](https://md.research.elamur.ai/free/en/papers/koehn-2004-statistical-significance-tests-machine-translation/). In Proceedings of the 2004 Conference on Empirical Methods in Natural Language Processing.
17. Philipp Koehn. 2009. Statistical Machine Translation. Cambridge University Press. [find](https://scholar.google.com/scholar?q=Philipp+Koehn%2E+2009%2E+Statistical+Machine+Translation%2E+Cambridge+University+Press%2E)
18. Philipp Koehn, Alexandra Birch, and Ralf Steinberger. 2009. 462 machine translation systems for Europe. Proceedings of MT Summit XII, pages 65–72. [find](https://scholar.google.com/scholar?q=Philipp+Koehn%2C+Alexandra+Birch%2C+and+Ralf+Steinberger%2E+2009%2E+462+machine+translation+systems+for+Europe%2E+Proceedings+of+MT+Summit+XII%2C+pages+65%E2%80%9372%2E)
19. Philipp Koehn, Hieu Hoang, Alexandra Birch, Chris Callison-Burch, Marcello Federico, Nicola Bertoldi, Brooke Cowan, Wade Shen, Christine Moran, Richard Zens, et al. 2007. Moses: Open source toolkit for statistical machine translation. In Proceedings of the 45th Annual Meeting of the Association for… [find](https://scholar.google.com/scholar?q=Philipp+Koehn%2C+Hieu+Hoang%2C+Alexandra+Birch%2C+Chris+Callison%2DBurch%2C+Marcello+Federico%2C+Nicola+Bertoldi%2C+Brooke+Cowan%2C+Wade+Shen%2C+Christine+Moran%2C+Richard+Zens%2C+et+al%2E+2007%2E+Moses%3A+Open+source+toolkit+fo)
20. Philipp Koehn and Jean Senellart. 2010a. [Convergence of translation memory and statistical machine translation](https://md.research.elamur.ai/free/en/papers/koehn-2010-convergence-translation-memory-statistical-machine/). In Proceedings of AMTA Workshop on MT Research and the Translation Industry, pages 21–31.
21. Philipp Koehn and Jean Senellart. 2010b. Fast approximate string matching with suffix arrays and a* parsing. In Proceedings of the ninth Conference of the Association for Machine Translation in the Americas. [find](https://scholar.google.com/scholar?q=Philipp+Koehn+and+Jean+Senellart%2E+2010b%2E+Fast+approximate+string+matching+with+suffix+arrays+and+a%2A+parsing%2E+In+Proceedings+of+the+ninth+Conference+of+the+Association+for+Machine+Translation+in+the+Am)
22. Alon Lavie and Abhaya Agarwal. 2007. [METEOR: An automatic metric for MT evaluation with high levels of correlation with human judgments](https://md.research.elamur.ai/free/en/papers/lavie-2007-meteor-automatic-metric-mt-evaluation/). In Proceedings of the Second Workshop on Statistical Machine Translation, pages 228–231.
23. Vladimir Levenshtein. 1966. Binary codes capable of correcting deletions, insertions, and reversals. Soviet Physics Doklady, 10(8):707–710. [find](https://scholar.google.com/scholar?q=Vladimir+Levenshtein%2E+1966%2E+Binary+codes+capable+of+correcting+deletions%2C+insertions%2C+and+reversals%2E+Soviet+Physics+Doklady%2C+10%288%29%3A707%E2%80%93710%2E)
24. Gonzalo Navarro. 2001. A guided tour to approximate string matching. ACM computing surveys (CSUR), 33(1):31–88. [find](https://scholar.google.com/scholar?q=Gonzalo+Navarro%2E+2001%2E+A+guided+tour+to+approximate+string+matching%2E+ACM+computing+surveys+%28CSUR%29%2C+33%281%29%3A31%E2%80%9388%2E)
25. John Ortega, Felipe Sánchez-Martınez, and Mikel Forcada. 2016. Fuzzy-match repair using black-box machine translation systems: what can be expected? In Proceedings of the 2016 Conference of the Association for Machine Translation in the Americas, pages 27–39. [find](https://scholar.google.com/scholar?q=John+Ortega%2C+Felipe+Sa%CC%81nchez%2DMart%C4%B1nez%2C+and+Mikel+Forcada%2E+2016%2E+Fuzzy%2Dmatch+repair+using+black%2Dbox+machine+translation+systems%3A+what+can+be+expected%3F+In+Proceedings+of+the+2016+Conference+of+the+Assoc)
26. Kishore Papineni, Salim Roukos, Todd Ward, and Wei-Jing Zhu. 2002. [BLEU: a method for automatic evaluation of machine translation](https://md.research.elamur.ai/free/en/papers/papineni-2002-bleu-method-automatic-evaluation-machine/). In Proceedings of the 40th Annual Meeting of the Association for Computational Linguistics, pages 311–318.
27. Caroline Rossi and Jean-Pierre Chevrot. 2019. Uses and perceptions of machine translation at the European Commission. Journal of Specialised Translation, 31:177–200. [find](https://scholar.google.com/scholar?q=Caroline+Rossi+and+Jean%2DPierre+Chevrot%2E+2019%2E+Uses+and+perceptions+of+machine+translation+at+the+European+Commission%2E+Journal+of+Specialised+Translation%2C+31%3A177%E2%80%93200%2E)
28. Michel Simard and Pierre Isabelle. 2009. Phrase-based machine translation in a computer-assisted translation environment. In Proceedings of MT Summit XII, pages 120–127. [find](https://scholar.google.com/scholar?q=Michel+Simard+and+Pierre+Isabelle%2E+2009%2E+Phrase%2Dbased+machine+translation+in+a+computer%2Dassisted+translation+environment%2E+In+Proceedings+of+MT+Summit+XII%2C+pages+120%E2%80%93127%2E)
29. Michel Simard and Philippe Langlais. 2001. Sub-sentential exploitation of translation memories. In Machine Translation Summit 8, pages 335–339. [find](https://scholar.google.com/scholar?q=Michel+Simard+and+Philippe+Langlais%2E+2001%2E+Sub%2Dsentential+exploitation+of+translation+memories%2E+In+Machine+Translation+Summit+8%2C+pages+335%E2%80%93339%2E)
30. Matthew Snover, Bonnie Dorr, Richard Schwartz, Linnea Micciulla, and John Makhoul. 2006. [A study of translation edit rate with targeted human annotation](https://md.research.elamur.ai/free/en/papers/snover-2006-study-translation-edit-rate-targeted/). In Proceedings of the 2006 Conference of the Association for Machine Translation in the Americas, pages 223–231.
31. Ralf Steinberger, Andreas Eisele, Szymon Klocek, Spyridon Pilos, and Patrick Schlüter. 2013. [DGT-TM: A freely available translation memory in 22 languages](https://md.research.elamur.ai/free/en/papers/steinberger-2012-dgt-tm-freely-available-translation/). Computing Research Repository, arXiv:1309.5226.
32. Tom Vanallemeersch and Vincent Vandeghinste. 2015. Assessing linguistically aware fuzzy matching in translation memories. In Proceedings of the 18th Annual Conference of the European Association for Machine Translation, pages 153–160. [find](https://scholar.google.com/scholar?q=Tom+Vanallemeersch+and+Vincent+Vandeghinste%2E+2015%2E+Assessing+linguistically+aware+fuzzy+matching+in+translation+memories%2E+In+Proceedings+of+the+18th+Annual+Conference+of+the+European+Association+for+M)
33. Kun Wang, Chengqing Zong, and Keh-Yih Su. 2013. Integrating translation memory into phrase-based machine translation during decoding. In Proceedings of the 51st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long papers), pages 11–21. [find](https://scholar.google.com/scholar?q=Kun+Wang%2C+Chengqing+Zong%2C+and+Keh%2DYih+Su%2E+2013%2E+Integrating+translation+memory+into+phrase%2Dbased+machine+translation+during+decoding%2E+In+Proceedings+of+the+51st+Annual+Meeting+of+the+Association+for+C)
34. Jingyi Zhang, Masao Utiyama, Eiichro Sumita, Graham Neubig, and Satoshi Nakamura. 2018. [Guiding neural machine translation with retrieved translation pieces](https://md.research.elamur.ai/free/en/papers/zhang-2018-guiding-neural-machine-translation-retrieved/). Computing Research Repository, arXiv:1804.02559.

*Updated 24 October 2026*