← heapsort-ai

Cultural Bias

3 items

RESEARCHarXiv CS.CL·4/6/2026

Pragmatics Meets Culture: Culturally-adapted Artwork Description Generation and Evaluation

Este artigo apresenta a tarefa de geração de descrições de arte culturalmente adaptadas para combater o viés cultural em modelos de linguagem na geração de texto aberto. Ele propõe um framework de avaliação baseado em perguntas e respostas culturalmente fundamentadas, mostrando que um modelo de locutor pragmático melhora significativamente a compreensão do ouvinte.

28
RESEARCHarXiv CS.CL·26d ago

Mitigating Cross-Lingual Cultural Inconsistencies in LLMs via Consensus-Driven Preference Optimisation

Multilingual large language models (MLLMs) often exhibit inconsistent behavior regarding cultural identity when the prompt's language changes. Researchers introduce a new metric, Singleton Fleiss's "k_S", and a consensus-driven alignment framework, C-3PO, to mitigate these cross-lingual cultural inconsistencies, achieving significant improvements.

27
RESEARCHarXiv CS.CL·8d ago

When English Rewrites Local Knowledge: Global Narrative Dominance in Large Language Models

This research paper investigates global narrative dominance in Large Language Models (LLMs), where local cultural knowledge is often overshadowed by global narratives. It introduces the CulturalNB dataset for Bengali cultural contexts and demonstrates that questions asked in English tend to increase global substitution and institutional framing, reducing local perspective coverage.

27