arXiv cs.CV
· Papers
MUSON: A Reasoning-oriented Multimodal Dataset for Socially Compliant Navigation in Urban Environments
arXiv:2512.22867v2 Announce Type: replace Abstract: Socially compliant navigation requires structured reasoning about dynamic pedestrians and physical constraints to ensure safe and interpretable decisions. Vision-language models (VLMs) provide a promising foundation for this task because they can integrate visual obse