Editorial: Multi-modal learning with large-scale models

The integration of multi-modal learning with large-scale models has become a transformative 9 force in the fields of artificial intelligence and neurorobotics. Human perception naturally relies on the 10 seamless fusion of various sensory inputs-visual, auditory, tactile, and beyond-to navigate and 11 understand complex environments. Replicating this holistic capability in intelligent systems has 12 historically been constrained by computational limitations and the difficulty of aligning heterogeneous 13 data. However, the advent of large-scale models has shifted the paradigm, offering unprecedented 14 capacity to process, align, and fuse multi-modal data. This Research Topic, "Multi-modal Learning 15 with Large-scale Models," aims to explore the cutting edge of these architectures, emphasizing their 16 applications across robotic perception, autonomous navigation, human-robot interaction, and creative 17 generation. The seven articles gathered in this collection illustrate how multi-modal large-scale models 18 are bridging the gap between isolated data streams and unified machine cognition.