Eduard J. Burton. Cross-Modal World Modeling With HY-Himmel: Unifying Video, Text, and Sensor Streams for Embodied AI. CSET 2025, 3.