RAMMM: A Rapid Attention-Based Multimodal Modification Model for Massive Image Generation | IEEE Conference Publication | IEEE Xplore