[2506.18199] Prompt Engineering Techniques for Mitigating Cultural Bias Against Arabs and Muslims in Large Language Models: A Systematic Review

View the PDF file from the paper entitled Immediate Engineering Technologies to alleviate Cultural Bias against Arabs and Muslims in the Long Language Models: Systematic Review, by Busheri Asseri and 2 other authors

No pdf is available, click to view other formats

a summary:Language models have shown noticeable capabilities in various fields, however concerns about cultural bias – especially towards Arabs and Muslims – constitute great moral challenges by perpetuating harmful stereotypes and marginalization. Despite the increasing recognition of the bias in LLMS, the immediate engineering strategies that especially address Arab and Muslim representation remain. This mixed systematic review is looking at such technologies, as it provides evidence based on researchers and practitioners. After Prisma and the methodology of the methodological review of Kitchenham, we analyzed 8 experimental studies published between 2021-2024 an investigation into the bias mitigating strategies. The results we find reveals five basic engineering methods: cultural student, emotional preparation, self -navigation techniques, organized multi -step pipelines, and improved continuous demands for female teachers. Although all methods show the possibility of reducing bias, the event varies greatly through studies and types of bias. Evidence indicates that some types of prejudice may be more resistant to relieving the list based on others. Multi -steps organized pipelines showed the highest overall effectiveness, achieving up to 87.7 % in bias, although they require greater technical experience. Cultural claim provides wider access with great effectiveness. These results emphasize access to immediate engineering to alleviate cultural bias without the need to reach the parameters of the model. The limited number of studies that have been determined by a large research gap in this critical field is highlighted. Future research should focus on developing culturally adaptive debris, creating Arab and Muslim evaluation resources, and integrating unique engineering with supplementary supplementary methods to address deeper stereotypes while maintaining the benefit of the model.