Option-aware Temporally Abstracted Value for Offline Goal-Conditioned Reinforcement Learning | Read Paper on Bytez