LAGO: Few-shot Crosslingual Embedding Inversion Attacks via Language Similarity-Aware Graph Optimization

TL;DR

LAGO introduces a graph-based optimization method that leverages language similarities to enhance few-shot cross-lingual embedding inversion attacks, revealing privacy vulnerabilities in multilingual NLP systems.

Contribution

It models linguistic relationships via a graph framework, generalizing prior methods and improving attack transferability with limited data.

Findings

LAGO outperforms baselines with 10-20% higher Rouge-L scores.

Language similarity significantly impacts attack transferability.

The approach is effective across multiple languages and embedding models.

Abstract

We propose LAGO - Language Similarity-Aware Graph Optimization - a novel approach for few-shot cross-lingual embedding inversion attacks, addressing critical privacy vulnerabilities in multilingual NLP systems. Unlike prior work in embedding inversion attacks that treat languages independently, LAGO explicitly models linguistic relationships through a graph-based constrained distributed optimization framework. By integrating syntactic and lexical similarity as edge constraints, our method enables collaborative parameter learning across related languages. Theoretically, we show this formulation generalizes prior approaches, such as ALGEN, which emerges as a special case when similarity constraints are relaxed. Our framework uniquely combines Frobenius-norm regularization with linear inequality or total variation constraints, ensuring robust alignment of cross-lingual embedding spaces even with extremely limited data (as few as 10 samples per language). Extensive experiments across multiple languages and embedding models demonstrate that LAGO substantially improves the transferability of attacks with 10-20% increase in Rouge-L score over baselines. This work establishes language similarity as a critical factor in inversion attack transferability, urging renewed focus on language-aware privacy-preserving multilingual embeddings.