arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2405.20089cs.CL
收藏

微调悖论:在不牺牲 LLM 能力的情况下提升翻译质量

The Fine-Tuning Paradox: Boosting Translation Quality Without Sacrificing LLM Abilities

Amazon AI(亚马逊人工智能);University of Amsterdam(阿姆斯特丹大学)

David Stap, Eva Hasler, Bill Byrne, Christof Monz, Ke Tran

AI总结:

本文评估 LLaMA 与 Falcon 微调后的翻译能力,发现质量提升伴随可引导性、少样本技术翻译和文档级翻译退化,而加入单语数据可兼顾能力保持与质量增强。

摘要:

为机器翻译微调大型语言模型(LLM)已在整体翻译质量方面展现出改进。然而,目前尚不清楚微调会对神经机器翻译模型所不具备的理想 LLM 行为产生什么影响,例如可引导性、内在的文档级翻译能力,以及生成不那么逐字直译的译文的能力。我们对 LLaMA 和 Falcon 系列模型进行了广泛的翻译评估,模型规模从 70 亿到 650 亿参数不等。结果表明,尽管微调提升了 LLM 的总体翻译质量,但若干能力出现退化。具体而言,我们观察到模型在语体正式度引导、通过少样本示例生成技术类译文以及执行文档级翻译方面的能力下降。另一方面,我们观察到模型在平行数据上微调后会生成不那么逐字直译的译文。我们表明,通过将单语数据纳入微调数据,可以在保持这些能力的同时提升整体翻译质量。我们的发现强调,需要能够保留 LLM 在机器翻译中优势的微调策略。

英文摘要:

Fine-tuning large language models (LLMs) for machine translation has shown improvements in overall translation quality. However, it is unclear what is the impact of fine-tuning on desirable LLM behaviors that are not present in neural machine translation models, such as steerability, inherent document-level translation abilities, and the ability to produce less literal translations. We perform an extensive translation evaluation on the LLaMA and Falcon family of models with model size ranging from 7 billion up to 65 billion parameters. Our results show that while fine-tuning improves the general translation quality of LLMs, several abilities degrade. In particular, we observe a decline in the ability to perform formality steering, to produce technical translations through few-shot examples, and to perform document-level translation. On the other hand, we observe that the model produces less literal translations after fine-tuning on parallel data. We show that by including monolingual data as part of the fine-tuning data we can maintain the abilities while simultaneously enhancing overall translation quality. Our findings emphasize the need for fine-tuning strategies that preserve the benefits of LLMs for machine translation.

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。

↑