Attaching a VAE novelty detector to the DINO-WM world model and penalizing out-of-distribution predicted states in CEM planning lowers Chamfer distance on small-data robot manipulation benchmarks.
Deep learning, reinforcement learning, and world models,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.RO 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Bounding Distributional Shifts in World Modeling through Novelty Detection
Attaching a VAE novelty detector to the DINO-WM world model and penalizing out-of-distribution predicted states in CEM planning lowers Chamfer distance on small-data robot manipulation benchmarks.