arXiv:2601.21700v3 Announce Type: replace Abstract: Large Language Models (LLMs) increasingly support culturally sensitive decision making, yet often exhibit misalignment due to skewed pretraining data and the absence of structured value representations. Existing methods can steer outputs, but often lack demographic grounding and treat values as independent, unstructured signals, reducing consistency and interpretability. We propose OG-MAR, an Ontology-Guided Multi-Agent Reasoning framework. OG-MAR summarizes respondent-specific values from the World Values Survey (WVS) and constructs a global
The growing deployment of LLMs in diverse global contexts highlights the urgent need to address cultural biases and align AI behavior with varied human values.
This research addresses a fundamental limitation in current AI systems, paving the way for more globally acceptable and ethically reliable AI applications, particularly in sensitive domains.
The ability to integrate structured value representations directly into LLM architectures marks a significant step towards creating AI that is not merely accurate but also culturally intelligent and adaptable.
- · AI developers in non-Western markets
- · International organizations
- · Users in diverse cultural contexts
- · Ethical AI research
- · Monocultural AI developers
- · AI systems with unaddressed biases
- · Global platforms lacking cultural sensitivity
Increased trust and adoption of advanced AI systems in diverse cultural settings.
Development of new AI applications tailored to specific cultural norms and values, potentially fostering local AI economies.
The emergence of 'cultural AI auditors' or new regulatory frameworks focused on value alignment and bias mitigation in AI.
This signal links to a primary source. Continuum Brief monitors and indexes it as part of the live intelligence stream — we do not republish source content.
Read at arXiv cs.CL