Evaluating Knowledge-based Cross-lingual Inconsistency in Large Language Models

TL;DR

This paper examines cross-lingual inconsistencies in large language models, proposing new evaluation metrics to measure their semantic, accuracy, and timeliness consistency across languages, aiming to improve multilingual robustness.

Contribution

It introduces an innovative evaluation framework using LaBSE and new metrics for comprehensive assessment of cross-lingual consistency in LLMs.

Findings

LLMs exhibit significant cross-lingual inconsistencies.

The proposed metrics effectively quantify semantic, accuracy, and timeliness inconsistencies.

Findings suggest areas for improving multilingual capabilities in LLMs.

Abstract

This paper investigates the cross-lingual inconsistencies observed in Large Language Models (LLMs), such as ChatGPT, Llama, and Baichuan, which have shown exceptional performance in various Natural Language Processing (NLP) tasks. Despite their successes, these models often exhibit significant inconsistencies when processing the same concepts across different languages. This study focuses on three primary questions: the existence of cross-lingual inconsistencies in LLMs, the specific aspects in which these inconsistencies manifest, and the correlation between cross-lingual consistency and multilingual capabilities of LLMs.To address these questions, we propose an innovative evaluation method for Cross-lingual Semantic Consistency (xSC) using the LaBSE model. We further introduce metrics for Cross-lingual Accuracy Consistency (xAC) and Cross-lingual Timeliness Consistency (xTC) to comprehensively assess the models' performance regarding semantic, accuracy, and timeliness inconsistencies. By harmonizing these metrics, we provide a holistic measurement of LLMs' cross-lingual consistency. Our findings aim to enhance the understanding and improvement of multilingual capabilities and interpretability in LLMs, contributing to the development of more robust and reliable multilingual language models.