一句话改全球：让大模型跨语言知识同步更高效
Edit Once, Update Everywhere: A Simple Framework for Cross-Lingual Knowledge Synchronization in LLMs
- 用双阶段框架实现跨语言知识更新，先指令微调再偏好优化
- 在双语和多语言评测上平均提升8.19%，单语性能不下降
- 适合需要多语言知识维护的模型应用者，如国际客服系统
知识编辑可实现大语言模型对新信息或修正的高效适应，而无需全量重训练。然而，现有方法多聚焦单一语言或基础多语言编辑，难以实现真正的跨语言知识同步。为此，我们提出一种简单且高效的前沿方案——跨语言知识民主编辑（X-KDE），旨在将主语言的知识有效传播至其他语言。该方案包含两个阶段：(i) 跨语言编辑指令微调（XE-IT），在精心构建的平行数据集上微调模型，修改特定知识同时保留无关信息；(ii) 目标语言偏好优化（TL-PO），采用先进优化技术确保多语言间一致性，促进更新传递。此外，我们还构建了一个高质量的跨语言数据集，专门用于增强跨语言知识迁移。在Bi-ZsRE和MzsRE基准上的大量实验表明，X-KDE显著提升了跨语言性能，平均提升8.19%，同时在单语言设置中保持高精度。
原文摘要 · Abstract (English)
Knowledge editing allows for efficient adaptation of large language models (LLMs) to new information or corrections without requiring full retraining. However, prior methods typically focus on either single-language editing or basic multilingual editing, failing to achieve true cross-linguistic knowledge synchronization. To address this, we present a simple and practical state-of-the-art (SOTA) recipe Cross-Lingual Knowledge Democracy Edit (X-KDE), designed to propagate knowledge from a dominant language to other languages effectively. Our X-KDE comprises two stages: (i) Cross-lingual Edition Instruction Tuning (XE-IT), which fine-tunes the model on a curated parallel dataset to modify in-scope knowledge while preserving unrelated information, and (ii) Target-language Preference Optimization (TL-PO), which applies advanced optimization techniques to ensure consistency across languages, fostering the transfer of updates. Additionally, we contribute a high-quality, cross-lingual dataset, specifically designed to enhance knowledge transfer across languages. Extensive experiments on the Bi-ZsRE and MzsRE benchmarks show that X-KDE significantly enhances cross-lingual performance, achieving an average improvement of +8.19%, while maintaining high accuracy in monolingual settings.
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。