一作ACL 2026Findings· Findings of ACL
Distilling the Essence, Discarding the Dross: Improving Fairness in Multimodal Large Language Models via Historical Reflection-Guided Prompt Optimization
Juncheng Hu · Jiming Yu · Rui Song · Kedi Lyu · Yingji Li · Zheli Liu
摘要
ocial bias in Multimodal Large Language Models (MLLMs) has become an increasingly important concern. Prompt-based approaches offer a lightweight solution for debiasing; however, existing methods rely heavily on handcrafted prompts that are brittle, highly context-sensitive, and difficult to gener...