Miao, R., Lu, X., Tian, C., Yu, B. and Duan, Z. 2026. T4NMTD: Transition-Centric Reinforcement Learning for Non-Markovian Task Decomposition. Proceedings of the AAAI Conference on Artificial Intelligence. 40, 29 (Mar. 2026), 24388-24395. DOI:https://doi.org/10.1609/aaai.v40i29.39620.