Metraipolis › archive › dc_c80af0c8e8a7
Rewrite What Matters: Adaptive Multilingual Query Rewriting for Reasoning via Agentic Reinforcement Learning — arXiv:2610.04899
- Source
- arXiv
- Original
- https://arxiv.org/abs/2610.04899
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-06T05:28:10.473Z
- SHA-256 of the archived text
d1e1147263bda35c065194720512e847aa8522db976810b6ec42283eef2672c3- Stable URL of this page
https://metraipolis.com/documento/dc_c80af0c8e8a7
How to cite
Rewrite What Matters: Adaptive Multilingual Query Rewriting for Reasoning via Agentic Reinforcement Learning — arXiv:2610.04899 — arXiv, https://arxiv.org/abs/2610.04899 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_c80af0c8e8a7 (Metraipolis archive dc_c80af0c8e8a7, sha256 d1e1147263bda35c065194720512e847aa8522db976810b6ec42283eef2672c3)
BibTeX
@misc{dcc80af0c8e8a7,
title = {Rewrite What Matters: Adaptive Multilingual Query Rewriting for Reasoning via Agentic Reinforcement Learning — arXiv:2610.04899},
howpublished = {arXiv},
url = {https://arxiv.org/abs/2610.04899},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_c80af0c8e8a7; sha256 d1e1147263bda35c065194720512e847aa8522db976810b6ec42283eef2672c3},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv, https://arxiv.org/abs/2610.04899. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
In multilingual scenarios, queries with equivalent semantics but in different languages could guide the model into different reasoning trajectories, leading to performance disparities. To mitigate this gap, previous studies typically apply a one-size-fits-all query rewriting strategy, such as translation, which overlooks the fact that different scenarios require diverse types of semantic transformations. In this paper, we propose mRewriter-R1, an agentic multilingual query rewriting framework with reinforcement learning. Unlike single-turn rewriting, mRewriter-R1 formulates multilingual query rewriting as a multi-turn sequential decision-making process, where the model dynamically performs multi-aspect optimization through adaptive operator selection. Experimental results demonstrate that mRewriter-R1 outperforms all strong multilingual rewriting baselines on different large reasoning backbones. Further analyses show that the learned policy can adaptively decide on rewriting operators according to query characteristics, exhibiting strong generalization ability across diverse reasoning tasks, and plug-and-play compatibility with heterogeneous reasoning language models. Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.CL, nuovi lavori del giorno). Categorie arXiv: Computation and Language (cs.CL). Pagina del lavoro: https://arxiv.org/abs/2610.04899 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-06T05:24:20.285Z dalla pagina https://arxiv.org/list/cs.CL/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).