MLLM-LLaVA-FL: Multimodal Large Language Model Assisted Federated Learning | IEEE Conference Publication | IEEE Xplore