STRIPS-WM induces an abstract transition graph from images, learns latent binary predicates and one grounded operator per action, then distills the predicates into a visual encoder for classical planning from novel start and goal images.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.RO 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
STRIPS-WM: Learning Grounded Propositional STRIPS-style World Models from Images
STRIPS-WM induces an abstract transition graph from images, learns latent binary predicates and one grounded operator per action, then distills the predicates into a visual encoder for classical planning from novel start and goal images.