Omni-modal decomposition autoencoders learn full-stack wearable disentangled representations

Part of

RIS-Assisted Semantic Channel Optimization