作者位次 #2/7ACL 2026Findings· Findings of ACL
Beyond Cross-Modal Alignment: Measuring and Leveraging Modality Gap in Vision-Language Models
Hanqi Yan · Xiangxiang Cui · Lu Yin · Jindong Gu · Paul Pu Liang · Yulan He · Yifei Wang
摘要
The success of vision-language models is primarily attributed to effective cross-modal alignment between vision and language. However, modality gaps persist even in well-aligned models and may be necessary for human perception, as evidenced by modality-specific phenomena such as visual texture an...