No tested generative video model achieves low rollout error, non-divergent temporal error, and real-time inference simultaneously for zero-shot predictive display in CARLA-simulated teleoperation.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.RO 1years
2026 1verdicts
UNVERDICTED 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
Towards Generative Predictive Display for Vision-Based Teleoperation: A Zero-Shot Benchmark of Off-the-Shelf Video Models
No tested generative video model achieves low rollout error, non-divergent temporal error, and real-time inference simultaneously for zero-shot predictive display in CARLA-simulated teleoperation.