TY  - RPRT
TI  - MAPS: Preserving Vision-Language Representations via Module-Wise Proximity Scheduling for Better Vision-Language-Action Generalization
AU  - Chengyue Huang
AU  - Mellon M. Zhang
AU  - Robert Azarcon
AU  - Glen Chou
AU  - Zsolt Kira
PY  - 2025
UR  - https://arxiv.org/abs/2511.19878
ID  - 2511.19878
ER  -