TY  - RPRT
TI  - Towards Fine-Grained Robustness: Attention-Guided Test-Time Prompt Tuning for Vision-Language Models
AU  - Jia-Wei Hai
AU  - Yijun Wang
AU  - Xiu-Shen Wei
PY  - 2026
UR  - https://arxiv.org/abs/2605.19956
ID  - 2605.19956
ER  -