Extending linear LAMs to model exogenous state shows standard reconstruction encodes future exogenous info in latent actions, while endogenous-focused spaces and auxiliary objectives like action-supervision enforce consistency across noise.
Self- supervised visual state representation learning for robotics from dynamic scenes
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Why Latent Actions Fail, and How to Prevent It
Extending linear LAMs to model exogenous state shows standard reconstruction encodes future exogenous info in latent actions, while endogenous-focused spaces and auxiliary objectives like action-supervision enforce consistency across noise.