TY  - RPRT
TI  - In-Context Collapse in Vision-Language Models and How to Mitigate it?
AU  - Mohammad Rostami
PY  - 2026
UR  - https://arxiv.org/abs/2608.02830
ID  - 2608.02830
ER  -