A 0.9B policy that couples action generation with latent future-state prediction in its own representation space, plus geometric supervision, reaches 98.5% LIBERO and 73% LIBERO-Plus.
π0.5: a vision-language-action model with open-world gener- alization
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
CONDITIONAL 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
SG-WAM: Self-Guided World Modeling in Geometry-Aware Policy Space
A 0.9B policy that couples action generation with latent future-state prediction in its own representation space, plus geometric supervision, reaches 98.5% LIBERO and 73% LIBERO-Plus.