Showing cs.CVShow all
2 papers · 1 filter
cs.CV2025
SocialNav-MoE: A Mixture-of-Experts Vision Language Model for Socially Compliant Navigation with Reinforcement Fine-Tuning
Tomohito Kawabata, Xinyu Zhang, Ling Xiao
For robots navigating in human-populated environments, safety and social compliance are equally critical, yet prior work has mostly emphasized safety. Socially compliant navigation…
cs.CV2025
MUSON: A Reasoning-oriented Multimodal Dataset for Socially Compliant Navigation in Urban Environments
Zhuonan Liu, Xinyu Zhang, Zishuo Wang +8
Socially compliant navigation requires structured reasoning about dynamic pedestrians and physical constraints to ensure safe and interpretable decisions. Vision-language models (V…