AI RESEARCH
Mitigating Cross-Lingual Cultural Inconsistencies in LLMs via Consensus-Driven Preference Optimisation
arXiv CS.CL
•
ArXi:2605.12515v1 Announce Type: new Despite their impressive capabilities, multilingual large language models (MLLMs) frequently exhibit inconsistent behaviour when the prompt's language changes. While such adaptation is generally desirable, it becomes a critical failure when a user's identity is explicitly defined. For instance, given a fixed British persona and an ambiguous everyday knowledge query about literature, the prompt's language frequently overwrites the system persona -- yielding Shakespeare in English but Cervantes in Spanish.