所选语言是跨语言真实性迁移所需的全部
Selected Languages are All You Need for Cross-lingual Truthfulness Transfer
AI总结：
本文提出一种名为FaMSS的跨语言真实性迁移方法，通过语言偏差和迁移贡献选择最优语言子集并采用翻译指令微调，有效减少了多语言表示差异并提升了LLMs的跨语言真实性。
摘要：
真实性是大型语言模型（LLMs）面临的一项关键挑战。尽管许多研究开发了多种真实性增强方法，但它们很少关注多语言场景下的真实性。同时，当代多语言对齐技术难以平衡众多语言，并且在不同语言之间经常表现出严重的真实性差距，尤其是那些与英语差异巨大的语言。在我们的工作中，我们将真实性评估扩展到多语言语境，并提出了一种实用的跨语言真实性迁移方法，称为事实感知多语言选择性协同（FaMSS）。FaMSS能够通过语言偏差和迁移贡献从所有测试语言中选择一个最优子集，然后采用翻译指令微调进行跨语言真实性迁移。实验结果表明，我们的方法能够有效减少多语言表示差异，并提升LLMs的跨语言真实性迁移。
英文摘要：
Truthfulness stands out as an essential challenge for Large Language Models (LLMs). Although many works have developed various ways for truthfulness enhancement, they seldom focus on truthfulness in multilingual scenarios. Meanwhile, contemporary multilingual aligning technologies struggle to balance numerous languages and often exhibit serious truthfulness gaps across different languages, especially those that differ greatly from English. In our work, we extend truthfulness evaluation to multilingual contexts and propose a practical method for cross-lingual truthfulness transfer called Fact-aware Multilingual Selective Synergy (FaMSS). FaMSS is able to select an optimal subset of all tested languages by language bias and transfer contributions, and then employ translation instruction tuning for cross-lingual truthfulness transfer. Experimental results demonstrate that our approach can effectively reduce the multilingual representation disparity and boost cross-lingual truthfulness transfer of LLMs.
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。
