Metraipolis › archive › dc_94ebf19283ff
PAMT: Process-Aligned Reinforcement Learning for Multi-Domain Machine Translation — arXiv:2608.03077
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.03077
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-31T12:30:55.825Z
- SHA-256 of the archived text
d3c967d426bb10a2044fc995337ef56cd4efc60c7fa11bbfe03f34cec1c685aa- Stable URL of this page
https://metraipolis.com/documento/dc_94ebf19283ff
How to cite
PAMT: Process-Aligned Reinforcement Learning for Multi-Domain Machine Translation — arXiv:2608.03077 — arXiv (en), https://arxiv.org/abs/2608.03077 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_94ebf19283ff (Metraipolis archive dc_94ebf19283ff, sha256 d3c967d426bb10a2044fc995337ef56cd4efc60c7fa11bbfe03f34cec1c685aa)
BibTeX
@misc{dc94ebf19283ff,
title = {PAMT: Process-Aligned Reinforcement Learning for Multi-Domain Machine Translation — arXiv:2608.03077},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.03077},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_94ebf19283ff; sha256 d3c967d426bb10a2044fc995337ef56cd4efc60c7fa11bbfe03f34cec1c685aa},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.03077. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Multi-domain machine translation (MDMT) requires more than fluent generation: it demands domain-sensitive translation decisions such as domain disambiguation, terminology control, and stylistic adaptation. Large reasoning models (LRMs) make such decisions explicit through intermediate translation steps, but our analysis across 15 domains and four translation directions shows that this explicit reasoning is double-edged: it improves long-form and high-difficulty translation, yet often drifts in terminology-intensive and stylistically constrained settings. We trace this failure to a credit-assignment bottleneck: existing methods optimize final outputs or coarse trajectories, but cannot identify which translation steps actually help the final translation. To address this, we propose PAMT, a process-aligned training framework that combines cold-start domain-aware Long-CoT supervision with reinforcement learning. PAMT uses sequence-level format and outcome rewards for the final translation, together with a step-level process reward that measures how much each explicit translation step increases the likelihood of the reference translation. Across two backbones, PAMT improves over base models, outperforms MT-specialized baselines on average, and remains competitive with strong LLMs/LRMs across in-domain, OOD, and multilingual settings. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-05 in cs.CL). Categorie arXiv: Computation and Language (cs.CL). Pagina del lavoro: https://arxiv.org/abs/2608.03077 — 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 2026-10-31T12:17:37.738Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-08-05?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).