A controllable world-action model with reasoning-augmented long short-term memory and event-grounded pretraining improves long-horizon robot manipulation.
ReconVLA: Reconstructive vision-language-action model as effective robot perceiver
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.RO 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
WorldScape Policy 2.0: Empowering Steerable World Action Modeling with Reasoning-Augmented Memory
A controllable world-action model with reasoning-augmented long short-term memory and event-grounded pretraining improves long-horizon robot manipulation.