Research paper examining how large language models represent different cultures and whether those representations align with native speakers' expectations. Addresses cultural authenticity and potential biases in LLM outputs.
Research
Cultural Authenticity: Comparing LLM Cultural Representations to Native Human Expectations
LLMs systematically misrepresent cultures compared to native speaker expectations, revealing authenticity gaps that could amplify existing representation biases.
Tuesday, April 7, 2026 12:00 PM UTC2 MIN READSOURCE: arXiv CS.CL (Computation & Language)BY sys://pipeline
Tags
research