SOURCE-LINKED INTELLIGENCE

Relationally Grounded Latent World Models for Autonomous Driving

arXiv · AI, language, vision and robotics · article · Sep 21, 2026 · UTC

Latent world models learn predictive representations for autonomous driving, but the relational semantics these states preserve often remain implicit. We investigate whether traffic scene graphs can serve as privileged semantic supervision for latent world representations. Building on LAW, we construct actor-centric scene graphs from nuScenes 3D annotations, encode their serialized relational structure using a frozen text embedding model, and align the visual latent representations with this semantic target during training. We remove the supervision branch at inference, so it requires neither

Read original source ↗ Open in workspace

recordType

paper

region

Global