AI RESEARCH
Belief-State RWKV for Reinforcement Learning under Partial Observability
arXiv CS.LG
•
ArXi:2604.09671v1 Announce Type: new We propose a stronger formulation of RL on top of RWKV-style recurrent sequence models, in which the fixed-size recurrent state is explicitly interpreted as a belief state rather than an opaque hidden vector. Instead of conditioning policy and value on a single summary h_t, we maintain a compact uncertainty-aware state b_t = (\mu_t, \Sigma_t) derived from RWKV-style recurrent statistics and let control depend on both memory and uncertainty.