Preprint
Aug 2026
EndoWAM: A Grounded World-Action Model for Generalizable Endoscopic Navigation
EndoWAM is presented, which is, to the authors' knowledge, the first WAM for generalizable robotic endoscopic navigation and introduces future grounding, which predicts task-relevant target regions in future observations from intermediate denoising features of a video world model.
Jinsong Lin, Zikang Pan, Wanhao Liu et al.
· 1 citation