SALAMANDRATA模型专为38种欧洲语言翻译优化，7B版本参与WMT25竞赛并公开发布。
From SALAMANDRA to SALAMANDRATA: BSC Submission for WMT25 General Machine Translation Shared Task
- 基于持续预训练与指令微调的两阶段训练，适配多语言翻译任务。
- 在WMT25共享任务中，7B版模型跨方向翻译表现优异，支持非欧洲语言扩展。
- 采用质量感知解码策略，适合多语言机器翻译研究与应用者使用。
本文介绍SALAMANDRATA系列模型，是SALAMANDRA LLMs（Gonzalez-Agirre et al., 2025）的改进版本，专为38种欧洲语言的翻译任务设计。SALAMANDRATA分为2B和7B参数量两种规模，均采用相同训练流程：第一阶段在平行语料上进行持续预训练，第二阶段在高质量指令数据上进行监督微调。本次参与WMT25通用机器翻译共享任务的BSC提交基于7B版本。我们首先将模型词汇表扩展以支持任务中新增的非欧洲语言，随后进行第二阶段持续预训练与微调，专门优化所有翻译方向的表现。解码阶段采用最小贝叶斯风险解码及分别基于COMET和COMET-KIWI的调优重排序策略。我们已在Hugging Face上公开发布2B和7B版本的SALAMANDRATA，以及更新的SALAMANDRATA-V2模型。
原文摘要 · Abstract (English)
In this paper, we present the SALAMANDRATA family of models, an improved iteration of SALAMANDRA LLMs (Gonzalez-Agirre et al., 2025) specifically trained to achieve strong performance in translation-related tasks for 38 European languages. SALAMANDRATA comes in two scales: 2B and 7B parameters. For both versions, we applied the same training recipe with a first step of continual pre-training on parallel data, and a second step of supervised fine-tuning on high-quality instructions. The BSC submission to the WMT25 General Machine Translation shared task is based on the 7B variant of SALAMANDRATA. We first adapted the model vocabulary to support the additional non-European languages included in the task. This was followed by a second phase of continual pre-training and supervised fine-tuning, carefully designed to optimize performance across all translation directions for this year's shared task. For decoding, we employed two quality-aware strategies: Minimum Bayes Risk Decoding and Tuned Re-ranking using COMET and COMET-KIWI respectively. We publicly release both the 2B and 7B versions of SALAMANDRATA, along with the newer SALAMANDRATA-V2 model, on Hugging Face1
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。