AI RESEARCH
Position: LLMs Must Use Functor-Based and RAG-Driven Bias Mitigation for Fairness
arXiv CS.CL
•
ArXi:2603.07368v1 Announce Type: new Biases in large language models (LLMs) often manifest as systematic distortions in associations between graphic attributes and professional or social roles, reinforcing harmful stereotypes across gender, ethnicity, and geography. This position paper advocates for addressing graphic and gender biases in LLMs through a dual-pronged methodology, integrating category-theoretic transformations and retrieval-augmented generation (RAG