9 citations · 16 across the 9 of their papers we have counts for
Showing 2026Show all
3 papers · 1 filter
cs.RO2026
Remembering What Matters for Long-Horizon Robotic World Models
Hao Wu, Tianyu Shao, Shengju Qian +10
Autoregressive robot video world models based on discrete visual tokens remain unreliable over long rollout horizons. Retaining the entire generated history does not necessarily im…
cs.CV2026
FutureWorlds: Learning Robotic World Models from Alternative Futures
Hao Wu, Shengju Qian, Weiyan Wang +5
Robotic world models predict action-conditioned future scenes, providing a foundation for understanding action outcomes. However, turning alternative predictions into useful learni…
cs.RO2026
RoboAlign-R1: Distilled Multimodal Reward Alignment for Robot Video World Models
Hao Wu, Yuqi Li, Yuan Gao +10
Existing robot video world models are typically trained with low-level objectives such as reconstruction and perceptual similarity, which are poorly aligned with the capabilities t…