arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2603.07372cs.CLcs.AIcs.LG
收藏

领域特定机器翻译低资源场景下的质量估计

Domain-Specific Quality Estimation for Machine Translation in Low-Resource Scenarios

Institute for People-Centred AI, University of Surrey(萨里大学以人为本人工智能研究所)

Namrata Patil Gurav, Akashdeep Ranu, Archchana Sindhujan, Diptesh Kanojia

AI总结:

本文提出ALOPE框架,通过低秩适应提升领域特定机器翻译低资源场景下的质量估计性能,尤其在语义复杂领域表现更优。

摘要:

质量估计(QE)对于在无参考设置下评估机器翻译质量至关重要,特别是在领域特定和低资源语言场景中。本文研究了英语到印地语机器翻译在四个领域(医疗、法律、旅游和一般)和五个语言对上的句子级QE。我们系统比较了零样本、少样本和指南锚定提示在选定的封闭权重和开放权重LLM上的表现。研究发现,封闭权重模型通过提示本身实现强大表现,但提示-only方法在开放权重模型中尤其脆弱,尤其是在高风险领域。为此,我们采用ALOPE,一种基于LLM的质量估计框架,它使用低秩适应并在选定的中间Transformer层上附加回归头。我们还扩展了ALOPE,结合最近提出的低秩乘法适应(LoRMA)。我们的结果表明,中间层适应一致地提高了QE性能,在语义复杂的领域中取得了收益,表明了在实际场景中更稳健QE的路径。我们公开发布了代码和领域特定QE数据集以支持进一步研究。

英文摘要:

Quality Estimation (QE) is essential for assessing machine translation quality in reference-less settings, particularly for domain-specific and low-resource language scenarios. In this paper, we investigate sentence-level QE for English to Indic machine translation across four domains (Healthcare, Legal, Tourism, and General) and five language pairs. We systematically compare zero-shot, few-shot, and guideline-anchored prompting across selected closed-weight and open-weight LLMs. Findings indicate that while closed-weight models achieve strong performance via prompting alone, prompt-only approaches remain fragile for open-weight models, especially in high-risk domains. To address this, we adopt ALOPE, a framework for LLM-based QE that uses Low-Rank Adaptation with regression heads attached to selected intermediate Transformer layers. We also extend ALOPE with recently proposed Low-Rank Multiplicative Adaptation (LoRMA). Our results show that intermediate-layer adaptation consistently improves QE performance, with gains in semantically complex domains, indicating a path toward more robust QE in practical scenarios. We release code and domain-specific QE datasets publicly to support further research.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑