Information-Time Proximal Policy Optimization
This paper proposes InfoPPO, which reparameterizes temporal progression using information density rather than raw token count, and extends performance-difference and policy-improvement analyses to the information-time MDP, deriving a policy-improvement lower bound when policy changes are regulated by information densit...
Yong-Cheng Zeng, Xin-Yu Cui, Yan Song et al.
· 0 citations