Multilingual LLMs display cross-lingual cultural inconsistency that a new metric quantifies and a consensus-driven preference optimization method reduces by up to 0.10 points.
Cultural Tendencies in Generative AI
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
citation-role summary
background 1
other 1
citation-polarity summary
years
2026 3representative citing papers
LLMs perceive cities through a culturally uneven baseline that favors Europe and Northern America over other regions in both open descriptions and structured judgments.
citing papers explorer
-
Mitigating Cross-Lingual Cultural Inconsistencies in LLMs via Consensus-Driven Preference Optimisation
Multilingual LLMs display cross-lingual cultural inconsistency that a new metric quantifies and a consensus-driven preference optimization method reduces by up to 0.10 points.
-
Large language models perceive cities through a culturally uneven baseline
LLMs perceive cities through a culturally uneven baseline that favors Europe and Northern America over other regions in both open descriptions and structured judgments.
- Agentivism: a learning theory for the age of artificial intelligence