TY  - RPRT
TI  - Belief-State RWKV for Reinforcement Learning under Partial Observability
AU  - Liu Xiao
PY  - 2026
UR  - https://arxiv.org/abs/2604.09671
ID  - 2604.09671
ER  -