TY  - RPRT
TI  - From Static Constraints to Dynamic Adaptation: Sample-Level Constraint Relaxation for Offline-to-Online Reinforcement Learning
AU  - Lipeng Zu
AU  - Yu Qian
AU  - Shayok Chakraborty
AU  - Xiaonan Zhang
PY  - 2026
UR  - https://arxiv.org/abs/2511.03828
ID  - 2511.03828
ER  -