Enhancing Stable Behavioral Imitation through Adaptive Reward Weighting in TD3-SAC-GAIL. JECIR. 2026;4(2):57-74. doi:10.63544/tvffe878