Search papers, labs, and topics across Lattice.
This paper introduces RAE-NWM, a navigation world model that operates in a dense visual representation space using DINOv2 features, addressing the limitations of VAE-based models that discard fine-grained spatial information. They find that DINOv2 features exhibit stronger linear predictability for action-conditioned transitions compared to VAE latent spaces. RAE-NWM employs a Conditional Diffusion Transformer with Decoupled Diffusion Transformer head (CDiT-DH) and a time-driven gating module to model continuous transitions, demonstrating improved structural stability, action accuracy, and downstream planning performance.
Ditching VAE bottlenecks for dense DINOv2 features unlocks more stable and accurate visual navigation world models.
Visual navigation requires agents to reach goals in complex environments through perception and planning. World models address this task by simulating action-conditioned state transitions to predict future observations. Current navigation world models typically learn state evolution under actions within the compressed latent space of a Variational Autoencoder, where spatial compression often discards fine-grained structural information and hinders precise control. To better understand the propagation characteristics of different representations, we conduct a linear dynamics probe and observe that dense DINOv2 features exhibit stronger linear predictability for action-conditioned transitions. Motivated by this observation, we propose the Representation Autoencoder-based Navigation World Model (RAE-NWM), which models navigation dynamics in a dense visual representation space. We employ a Conditional Diffusion Transformer with Decoupled Diffusion Transformer head (CDiT-DH) to model continuous transitions, and introduce a separate time-driven gating module for dynamics conditioning to regulate action injection strength during generation. Extensive evaluations show that modeling sequential rollouts in this space improves structural stability and action accuracy, benefiting downstream planning and navigation.
