Generating Adversarial Texts for Machine Translation via GRPO

• URL: http://arxiv.org/abs/2609.06048v1
• Date: Sat, 05 Sep 2026 12:06:59 GMT
• Title: Generating Adversarial Texts for Machine Translation via GRPO
• Abstract summary: We propose a scalable reinforcement-learning-based approach for rewriting existing source texts.<n>On WMT25, our approach substantially reduces average COMET translation quality from 0.63 to 0.48, while preserving grammaticality and readability.<n> Evaluations on the unseen WMT19-WMT24 benchmarks confirm that this behavior generalizes beyond the training data.
• Score: 9.016860740260679
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: As machine translation (MT) systems continue to improve, standard benchmarks become less informative for exposing remaining weaknesses. Traditional methods for creating challenging test sets rely on expensive manual creation or curation, while automated approaches struggle to produce sets with the necessary translation difficulty and linguistic diversity. We propose a scalable reinforcement-learning-based approach for rewriting existing source texts into instances that are more difficult to translate for MT systems. We fine-tune a large language model with Group Relative Policy Optimization (GRPO), using reward signals based on translation difficulty together with constraints for semantic similarity, grammaticality, and approximate length preservation. On WMT25, our approach substantially reduces average COMET translation quality from 0.63 to 0.48, while preserving grammaticality and readability, whereas the base model remains at 0.64. Evaluations on the unseen WMT19-WMT24 benchmarks confirm that this behavior generalizes beyond the training data, and human evaluation further shows that the rewrites substantially lower translation quality while incurring a moderate drop in naturalness and only a small change in grammaticality. We release our code to support reproducibility.

Related papers

• Unlocking Reasoning Capability on Machine Translation in Large Language Models [57.60641851466707]
  Reasoning-oriented large language models (RLMs) achieve strong gains on tasks such as mathematics and coding by generating explicit intermediate reasoning.<n>We systematically evaluate several open- and closed-weights RLMs on the WMT24++ benchmark.<n>We find that enabling explicit reasoning consistently degrades translation quality across languages and models.
  arXiv  Detail & Related papers  (2026-02-16T14:05:59Z)
• Beyond Literal Mapping: Benchmarking and Improving Non-Literal Translation Evaluation [57.11989521509119]
  We propose a novel agentic translation evaluation framework, centered by a reflective Core Agent that invokes specialized sub-agents.<n> Experimental results indicate the efficacy of RATE, achieving an improvement of at least 3.2 meta score compared with current metrics.
  arXiv  Detail & Related papers  (2026-01-12T09:03:42Z)
• Low-Resource English-Tigrinya MT: Leveraging Multilingual Models, Custom Tokenizers, and Clean Evaluation Benchmarks [6.177998679139308]
  Despite advances in Neural Machine Translation (NMT), low-resource languages like Tigrinya remain underserved.<n>This paper investigates transfer learning techniques using multilingual pretrained models to enhance translation quality for morphologically rich, low-resource languages.
  arXiv  Detail & Related papers  (2025-09-24T15:02:57Z)
• Languages Still Left Behind: Toward a Better Multilingual Machine Translation Benchmark [11.068031181100276]
  We study data in four languages (Asante Twi, Japanese, Jinghpaw, and South Azerbaijani)<n>We uncover critical shortcomings in the benchmark's suitability for truly multilingual evaluation.<n>We advocate for multilingual MT benchmarks that use domain-general and culturally neutral source texts.
  arXiv  Detail & Related papers  (2025-08-28T07:52:42Z)
• Estimating Machine Translation Difficulty [48.659971048116766]
  We formalize the task of translation difficulty estimation, defining a text's difficulty based on the expected quality of its translations.<n>We demonstrate the practical utility of difficulty estimators by using them to construct more challenging benchmarks for machine translation.<n>We release two improved models for difficulty estimation, Sentinel-src-24 and Sentinel-src-25.
  arXiv  Detail & Related papers  (2025-08-13T20:22:58Z)
• Do LLMs Understand Your Translations? Evaluating Paragraph-level MT with Question Answering [68.3400058037817]
  We introduce TREQA (Translation Evaluation via Question-Answering), a framework that extrinsically evaluates translation quality.<n>We show that TREQA is competitive with and, in some cases, outperforms state-of-the-art neural and LLM-based metrics in ranking alternative paragraph-level translations.
  arXiv  Detail & Related papers  (2025-04-10T09:24:54Z)
• A Data Selection Approach for Enhancing Low Resource Machine Translation Using Cross-Lingual Sentence Representations [0.4499833362998489]
  This study focuses on the case of English-Marathi language pairs, where existing datasets are notably noisy. To mitigate the impact of data quality issues, we propose a data filtering approach based on cross-lingual sentence representations. Results demonstrate a significant improvement in translation quality over the baseline post-filtering with IndicSBERT.
  arXiv  Detail & Related papers  (2024-09-04T13:49:45Z)
• BitextEdit: Automatic Bitext Editing for Improved Low-Resource Machine Translation [53.55009917938002]
  We propose to refine the mined bitexts via automatic editing. Experiments demonstrate that our approach successfully improves the quality of CCMatrix mined bitext for 5 low-resource language-pairs and 10 translation directions by up to 8 BLEU points.
  arXiv  Detail & Related papers  (2021-11-12T16:00:39Z)
• When Does Translation Require Context? A Data-driven, Multilingual Exploration [71.43817945875433]
  proper handling of discourse significantly contributes to the quality of machine translation (MT) Recent works in context-aware MT attempt to target a small set of discourse phenomena during evaluation. We develop the Multilingual Discourse-Aware benchmark, a series of taggers that identify and evaluate model performance on discourse phenomena.
  arXiv  Detail & Related papers  (2021-09-15T17:29:30Z)

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.