TY  - RPRT
TI  - Optimizing VLP-aligned Multimodal Intent Representation with Correct Visual Instantiation for Zero-Shot Composed Image Retrieval
AU  - Xuri Ge
AU  - Chunhao Wang
AU  - Junchen Fu
AU  - Haokun Wen
AU  - Zhiwei Xu
AU  - Ying Zhou
AU  - Zhumin Chen
AU  - Pengjie Ren
AU  - Zhaochun Ren
AU  - Xin Xin
PY  - 2026
UR  - https://arxiv.org/abs/2609.36946
ID  - 2609.36946
ER  -