TY  - RPRT
TI  - Lens: Bringing the Right Semantic Perspective into Focus for Training-Free Multimodal Representation Learning
AU  - Xinran Liu
AU  - Shouqian Shi
AU  - Yixian Chen
AU  - Ruizhi Chen
AU  - Xin-Wei Yao
AU  - Sheng Zhong
PY  - 2026
UR  - https://arxiv.org/abs/2609.20252
ID  - 2609.20252
ER  -