Rewrite to Translate, Translate to Reward: Reinforcement Learning for Source Rewriting in Machine Translation

• URL: http://arxiv.org/abs/2606.08011v2
• Date: Wed, 10 Jun 2026 14:48:43 GMT
• Title: Rewrite to Translate, Translate to Reward: Reinforcement Learning for Source Rewriting in Machine Translation
• Abstract summary: Rewriting source text with large language models before translation has been shown to improve machine translation quality.<n>We propose RLSR (Reinforcement Learning for Source Rewriting), a reinforcement learning framework that trains the rewriting model with a reward based on the downstream translation-quality improvement produced by each rewrite.
• Score: 43.662995515707735
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Rewriting source text with large language models (LLMs) before translation has been shown to improve machine translation (MT) quality. However, we find that prompt-based rewriting can degrade translation quality rather than improve it, particularly when smaller LLMs, such as 4B-parameter models, are used. We argue that this limitation stems from the difficulty of controlling rewriting behavior through natural-language prompts alone: a rewrite is useful only if it improves downstream translation, yet existing prompt-based methods do not explicitly optimize for this signal. To address this issue, we propose RLSR (Reinforcement Learning for Source Rewriting), a reinforcement learning framework that trains the rewriting model with a reward based on the downstream translation-quality improvement produced by each rewrite. Experiments across six MT systems and 16 language pairs show that our 4B RLSR-trained rewriting models significantly outperform both the no-rewriting baseline and prompt-based rewriting baselines at the same model scale, while remaining competitive with baselines that use a 235B LLM.

Related papers

• Generating Adversarial Texts for Machine Translation via GRPO [9.016860740260679]
  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.
  arXiv  Detail & Related papers  (2026-09-05T12:06:59Z)
• PEGRL: Improving Machine Translation by Post-Editing Guided Reinforcement Learning [54.19784655270799]
  We introduce textbfPEGRL, a textittwo-stage RL framework that uses post-editing as an auxiliary task to stabilize training and guide overall optimization.<n>Experiments on English$to$Finnish, English$to$Turkish, and English$leftrightarrow$Chinese show consistent gains over RL baselines.
  arXiv  Detail & Related papers  (2026-02-03T10:22:55Z)
• Automatic Input Rewriting Improves Translation with Large Language Models [24.855336353977382]
  Machine translation (MT) users rely on intuition that well-written text is easier to translate when using off-the-shelf MT systems.<n>We show that text simplification is the most effective MT-agnostic rewrite strategy.<n>Human evaluation confirms that simplified rewrites and their MT outputs both largely preserve the original meaning of the source and MT.
  arXiv  Detail & Related papers  (2025-02-23T18:56:56Z)
• RaFe: Ranking Feedback Improves Query Rewriting for RAG [83.24385658573198]
  We propose a framework for training query rewriting models free of annotations. By leveraging a publicly available reranker, oursprovides feedback aligned well with the rewriting objectives.
  arXiv  Detail & Related papers  (2024-05-23T11:00:19Z)
• Aligning Neural Machine Translation Models: Human Feedback in Training and Inference [27.84975767573212]
  Reinforcement learning from human feedback (RLHF) is a technique to improve the quality of the text generated by a language model. In machine translation (MT), where metrics trained from human annotations can readily be used as reward models, methods using minimum Bayes risk decoding and reranking have succeeded in improving the final quality of translation.
  arXiv  Detail & Related papers  (2023-11-15T17:21:58Z)
• Reinforced Self-Training (ReST) for Language Modeling [56.75447441157628]
  Reinforcement learning from human feedback (RLHF) can improve the quality of large language model's (LLM) outputs by aligning them with human preferences. We propose a simple algorithm for aligning LLMs with human preferences inspired by growing batch reinforcement learning (RL), which we call Reinforced Self-Training (ReST) Our results show that ReST can substantially improve translation quality, as measured by automated metrics and human evaluation on machine translation benchmarks in a compute and sample-efficient manner.
  arXiv  Detail & Related papers  (2023-08-17T14:12:48Z)
• Improving Non-autoregressive Translation Quality with Pretrained Language Model, Embedding Distillation and Upsampling Strategy for CTC [51.34222224728979]
  This paper introduces a series of innovative techniques to enhance the translation quality of Non-Autoregressive Translation (NAT) models. We propose fine-tuning Pretrained Multilingual Language Models (PMLMs) with the CTC loss to train NAT models effectively. Our model exhibits a remarkable speed improvement of 16.35 times compared to the autoregressive model.
  arXiv  Detail & Related papers  (2023-06-10T05:24:29Z)
• 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)
• The USYD-JD Speech Translation System for IWSLT 2021 [85.64797317290349]
  This paper describes the University of Sydney& JD's joint submission of the IWSLT 2021 low resource speech translation task. We trained our models with the officially provided ASR and MT datasets. To achieve better translation performance, we explored the most recent effective strategies, including back translation, knowledge distillation, multi-feature reranking and transductive finetuning.
  arXiv  Detail & Related papers  (2021-07-24T09:53:34Z)
• Rewriter-Evaluator Architecture for Neural Machine Translation [17.45780516143211]
  We present a novel architecture, Rewriter-Evaluator, for improving neural machine translation (NMT) models. It consists of a rewriter and an evaluator. At every pass, the rewriter produces a new translation to improve the past translation and the evaluator estimates the translation quality to decide whether to terminate the rewriting process. We conduct extensive experiments on two translation tasks, Chinese-English and English-German, and show that the proposed architecture notably improves the performances of NMT models.
  arXiv  Detail & Related papers  (2020-12-10T02:21:34Z)
• Enhanced back-translation for low resource neural machine translation using self-training [0.0]
  This work proposes a self-training strategy where the output of the backward model is used to improve the model itself through the forward translation technique. The technique was shown to improve baseline low resource IWSLT'14 English-German and IWSLT'15 English-Vietnamese backward translation models by 11.06 and 1.5 BLEUs respectively. The synthetic data generated by the improved English-German backward model was used to train a forward model which out-performed another forward model trained using standard back-translation by 2.7 BLEU.
  arXiv  Detail & Related papers  (2020-06-04T14:19:52Z)

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.