Intrigued by the claims of emergent reasoning capabilities in LLMs trained on general web corpora, in this paper, we set out to investigate their planning capabilities.

In this work, drawing insights from world models, we propose to instead learn a general representation that can adapt to be invariant or equivariant to different transformations by paying attention tocontext-- a memory module that tracks task-specificstates,actions,andfuturestates.

Three sampling techniques are presented for comparative investigations.

Much of the promise of these models relies on learned concepts being high quality .