Position: LLMs Must Use Functor-Based and RAG-Driven Bias Mitigation for Fairness

Ravi Ranjan, Utkarsh Grover, Agorista Polyzou

Biases in large language models (LLMs) often manifest as systematic distortions in associations between demographic attributes and professional or social roles, reinforcing harmful…