ASPECT:Analogical Semantic Policy Execution via Language Conditioned Transfer

ArXi:2604.08355v2 Announce Type: replace Reinforcement Learning (RL) agents often struggle to generalize knowledge to new tasks, even those structurally similar to ones they have mastered. Although recent approaches have attempted to mitigate this issue via zero-shot transfer, they are often constrained by predefined, discrete class systems, limiting their adaptability to novel or compositional task variations. We propose a significantly generalized approach, replacing discrete latent variables with natural language conditioning via a text-conditioned Variational Autoencoder.