Unified Multi-Modal Interactive & Reactive 3D Motion Generation via Rectified Flow

ArXi:2509.24099v3 Announce Type: replace Generating realistic, context-aware two-person motion conditioned on diverse modalities remains a fundamental challenge for graphics, animation and embodied AI systems. Real-world applications such as VR/AR companions, social robotics and game agents require models capable of producing coordinated interpersonal behaviour while flexibly switching between interactive and reactive generation. We