Graph-Attention Fusion With VAE Cross-Modal Mapping and Reinforcement-Learning Visualization for Real-Time AR. IJCAI 2025, 49 (14). https://doi.org/10.31449/inf.v49i14.11191.