Communications of the ACM
ACM Opinion
Large Language Models Associate Muslims with Violence
More research is urgently needed to better debias large language models. While applications are still in relatively early stages, these tasks may be influenced by Muslimviolence bias. Dialogue generated about Muslims may include excessive violence, or ne
Large language models, which are increasingly used in AI applications, display undesirable stereotypes, such as persistent associations between Muslims and violence.
Natural language processing (NLP) researchers must identify better ways to mitigate stereotypical bias against Muslims, as well as other social biases, that can be promoted by language models.
From Nature Machine Intelligence
View Full Article
No entries found