Edit Once, Update Everywhere: A Simple Framework for Cross-Lingual Knowledge Synchronization in LLMs

TL;DR

This paper introduces X-KDE, a simple yet effective framework for synchronizing knowledge updates across multiple languages in large language models, improving cross-lingual consistency without sacrificing monolingual accuracy.

Contribution

We propose a novel two-stage method, X-KDE, combining instruction tuning and optimization to enhance cross-lingual knowledge transfer in LLMs, supported by a new multilingual dataset.

Findings

X-KDE achieves +8.19% improvement on cross-lingual benchmarks.

The method maintains high monolingual accuracy.

Extensive experiments validate the effectiveness of the approach.

Abstract

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.