WM3C learns language-guided, composable causal components in a world model, with a block-wise identifiability guarantee, and demonstrates improved generalization to unseen simulated robot tasks.
However, this also adds some constraints and requirements on the number of tasks that combinedl 1 andl 2, which we are going to discuss later
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.AI 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Modeling Unseen Environments with Language-guided Composable Causal Components in Reinforcement Learning
WM3C learns language-guided, composable causal components in a world model, with a block-wise identifiability guarantee, and demonstrates improved generalization to unseen simulated robot tasks.