TY  - RPRT
TI  - GrndCtrl: Grounding World Models via Self-Supervised Reward Alignment
AU  - Haoyang He
AU  - Jay Patrikar
AU  - Dong-Ki Kim
AU  - Max Smith
AU  - Daniel McGann
AU  - Ali-akbar Agha-mohammadi
AU  - Shayegan Omidshafiei
AU  - Sebastian Scherer
PY  - 2026
UR  - https://arxiv.org/abs/2512.01952
ID  - 2512.01952
ER  -