Differences That Matter: Auditing Models for Capability Gap Discovery and Rectification

Qihao Liu, Chengzhi Mao, Yaojie Liu +2

Conventional evaluation methods for multimodal LLMs (MLLMs) lack interpretability and are often insufficient to fully disclose significant capability gaps across models. To address…