一作IROS 2026Poster· Generative and Predictive Models for Robot Behavior
GrndCtrl: Grounding World Models Via Self-Supervised Reward Alignment
Haoyang He · Jay Patrikar · Dong-Ki Kim · Max Smith · Daniel McGann · Ali-Akbar Agha-Mohammadi · Shayegan Omidshafiei · Sebastian Scherer
多模态对齐与偏好强化学习与后训练Vision-Based NavigationReinforcement LearningDeep Learning for Visual PerceptionOpenReview