BTBR: A Bayesian-Theory-Driven Probabilistic-Fuzzy Framework for Implicit Bias Removal in Large Language Models
Results demonstrate that combining probabilistic evidence with fuzzy degree modeling provides an effective and practical approach for mitigating implicit bias in large language models and Bayesian-Theory-based Bias Removal (BTBR) consistently reduces persona-induced performance gaps while preserving general reasoning a...
Yong-Xin Deng, Xiao-Yu Tan, Jing Pan et al.
· 4 citations