Learning to Translate Ambiguous Terminology by Preference Optimization on Post-Edits

• URL: http://arxiv.org/abs/2507.03580v2
• Date: Thu, 17 Jul 2025 10:42:09 GMT
• Title: Learning to Translate Ambiguous Terminology by Preference Optimization on Post-Edits
• Authors: Nathaniel Berger, Johannes Eschbach-Dymanus, Miriam Exel, Matthias Huck, Stefan Riezler,
• Abstract summary: In a corporate context, many examples of human post-edits of valid but incorrect terminology exist.<n>Our approach is based on preference optimization, using the term post-edit as the knowledge to be preferred.<n>We report results on English-German post-edited data and find that the optimal combination of supervised fine-tuning and preference optimization, with both term-specific and full sequence objectives, yields statistically significant improvements in term accuracy over a strong NMT baseline without significant losses in COMET score.
• Score: 10.580610673031073
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: In real world translation scenarios, terminology is rarely one-to-one. Instead, multiple valid translations may appear in a terminology dictionary, but correctness of a translation depends on corporate style guides and context. This can be challenging for neural machine translation (NMT) systems. Luckily, in a corporate context, many examples of human post-edits of valid but incorrect terminology exist. The goal of this work is to learn how to disambiguate our terminology based on these corrections. Our approach is based on preference optimization, using the term post-edit as the knowledge to be preferred. While previous work had to rely on unambiguous translation dictionaries to set hard constraints during decoding, or to add soft constraints in the input, our framework requires neither one-to-one dictionaries nor human intervention at decoding time. We report results on English-German post-edited data and find that the optimal combination of supervised fine-tuning and preference optimization, with both term-specific and full sequence objectives, yields statistically significant improvements in term accuracy over a strong NMT baseline without significant losses in COMET score. Additionally, we release test sets from our post-edited data and terminology dictionary.

Related papers

• Terminology-Aware Translation with Constrained Decoding and Large Language Model Prompting [11.264272119913311]
  We submit to the WMT 2023 terminology translation task. We adopt a translate-then-refine approach which can be domain-independent and requires minimal manual efforts. Results show that our terminology-aware model learns to incorporate terminologies effectively.
  arXiv  Detail & Related papers  (2023-10-09T16:08:23Z)
• Towards Effective Disambiguation for Machine Translation with Large Language Models [65.80775710657672]
  We study the capabilities of large language models to translate "ambiguous sentences" Experiments show that our methods can match or outperform state-of-the-art systems such as DeepL and NLLB in four out of five language directions.
  arXiv  Detail & Related papers  (2023-09-20T22:22:52Z)
• LEAPT: Learning Adaptive Prefix-to-prefix Translation For Simultaneous Machine Translation [6.411228564798412]
  Simultaneous machine translation is useful in many live scenarios but very challenging due to the trade-off between accuracy and latency. We propose a novel adaptive training policy called LEAPT, which allows our machine translation model to learn how to translate source prefixes and make use of the future context.
  arXiv  Detail & Related papers  (2023-03-21T11:17:37Z)
• DICTDIS: Dictionary Constrained Disambiguation for Improved NMT [50.888881348723295]
  We present DictDis, a lexically constrained NMT system that disambiguates between multiple candidate translations derived from dictionaries. We demonstrate the utility of DictDis via extensive experiments on English-Hindi and English-German sentences in a variety of domains including regulatory, finance, engineering.
  arXiv  Detail & Related papers  (2022-10-13T13:04:16Z)
• Rethink about the Word-level Quality Estimation for Machine Translation from Human Judgement [57.72846454929923]
  We create a benchmark dataset, emphHJQE, where the expert translators directly annotate poorly translated words. We propose two tag correcting strategies, namely tag refinement strategy and tree-based annotation strategy, to make the TER-based artificial QE corpus closer to emphHJQE. The results show our proposed dataset is more consistent with human judgement and also confirm the effectiveness of the proposed tag correcting strategies.
  arXiv  Detail & Related papers  (2022-09-13T02:37:12Z)
• Data-Driven Adaptive Simultaneous Machine Translation [51.01779863078624]
  We propose a novel and efficient training scheme for adaptive SimulMT. Our method outperforms all strong baselines in terms of translation quality and latency.
  arXiv  Detail & Related papers  (2022-04-27T02:40:21Z)
• DEEP: DEnoising Entity Pre-training for Neural Machine Translation [123.6686940355937]
  It has been shown that machine translation models usually generate poor translations for named entities that are infrequent in the training corpus. We propose DEEP, a DEnoising Entity Pre-training method that leverages large amounts of monolingual data and a knowledge base to improve named entity translation accuracy within sentences.
  arXiv  Detail & Related papers  (2021-11-14T17:28:09Z)
• Phrase-level Active Learning for Neural Machine Translation [107.28450614074002]
  We propose an active learning setting where we can spend a given budget on translating in-domain data. We select both full sentences and individual phrases from unlabelled data in the new domain for routing to human translators. In a German-English translation task, our active learning approach achieves consistent improvements over uncertainty-based sentence selection methods.
  arXiv  Detail & Related papers  (2021-06-21T19:20:42Z)
• Improving Lexically Constrained Neural Machine Translation with Source-Conditioned Masked Span Prediction [6.46964825569749]
  In this paper, we tackle a more challenging setup consisting of domain-specific corpora with much longer n-gram and highly specialized terms. To encourage span-level representations in generation, we additionally impose a source-sentence conditioned masked span prediction loss in the decoder. Experimental results on three domain-specific corpora in two language pairs demonstrate that the proposed training scheme can improve the performance of existing lexically constrained methods.
  arXiv  Detail & Related papers  (2021-05-12T08:11:33Z)
• Decoding Time Lexical Domain Adaptationfor Neural Machine Translation [7.628949147902029]
  Machine translation systems are vulnerable to domain mismatch, especially when the task is low-resource. We present two simple methods for improving translation quality in this particular setting.
  arXiv  Detail & Related papers  (2021-01-02T11:06:15Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.