aiqbench analysis
This research introduces RLSR, a reinforcement learning framework that automates source text rewriting for machine translation by using downstream translation quality as the reward signal. The approach eliminates the need for manual prompt tuning across different models, with experiments showing that a 4B parameter rewriting model outperforms baselines and rivals much larger 235B LLMs. For practitioners, this offers a scalable solution to enhance translation quality without relying on excessive compute resources or extensive hyperparameter optimization.
Rewrite to Translate, Translate to Reward: Reinforcement Learning for Source Rewriting in Machine Translation
arXiv:2606.08011v1 Announce Type: new Abstract: Although directly prompting off-the-shelf Large Language Models (LLMs) to generate meaning-preserving source rewrites can effectively enhance Machine Translation (MT) quality, doing so requires manually tuning prompts for different MT models. In this work, we propose RLSR (Reinforcement Learning for Source Rewriting), a novel RL-based framework for training a source rewriting model without tuning prompts for each MT model. RLSR optimizes the rewriting model by directly using the improvement in downstream translation quality yielded by each rewritten source as the reward. Extensive experiments across six MT models and 16 language pairs demonstrate that our 4B rewriting models trained via RLSR significantly outperform the no-rewriting baseline and existing same-scale prompt-based rewriting baselines, while achieving competitive performance against prompt-based baselines based on the 235B LLM.
Abstract:Although directly prompting off-the-shelf Large Language Models (LLMs) to generate meaning-preserving source rewrites can effectively enhance Machine Translation (MT) quality, doing so requires manually tuning prompts for different MT models. In this work, we propose RLSR (Reinforcement Learning for Source Rewriting), a novel RL-based framework for training a source rewriting model without tuning prompts for each MT model. RLSR optimizes the rewriting model by directly using the improvement in downstream translation quality yielded by each rewritten source as the reward. Extensive experiments across six MT models and 16 language pairs demonstrate that our 4B rewriting models trained via RLSR significantly outperform the no-rewriting baseline and existing same-scale prompt-based rewriting baselines, while achieving competitive performance against prompt-based baselines based on the 235B LLM.
Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
arXiv:2606.08011 [cs.CL]
(or arXiv:2606.08011v1 [cs.CL] for this version)
https://doi.org/10.48550/arXiv.2606.08011 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 220 words
Related coverage
Rewrite to Translate, Translate to Reward: Reinforcement Learning for Source Rewriting in Machine Translation
arXiv · Jun 11, 2026
Rewrite to Translate, Translate to Reward: Reinforcement Learning for Source Rewriting in Machine Translation
arXiv · Aug 5, 2026
Generating Adversarial Texts for Machine Translation via GRPO
arXiv · Sep 9, 2026
Podcast: The Future of AI: from Enterprise Adoption to Open Source Sovereignty
InfoQ · Sep 25, 2026