CausalWAM Isolates Action Causes in World Action Models via Sparse Routing
causal-intelligence · hf · 2026-11-02
A UIUC/UCSD paper tackles causal structure in World Action Models (WAMs): models fitted to demonstrations don't generally guarantee that actions are generated solely from the latent causes of those actions, as irrelevant latents can become entangled.
- Theory: Under a sparsity constraint, the latents used to generate actions are an invertible function of the action-relevant latents alone, with the true structure identifiable up to permutation—requiring neither sufficient temporal dependence assumptions nor statistical independence of action-relevant latents.
- Method: CausalWAM incorporates the constraint into a pretrained WAM by adding a sparse routing path to its action head while keeping the world-modeling objective unchanged, isolating identifiable causal structure.
More from Embodied
- Stanford's RoboArchitect Auto-Designs Robot Harnesses, Lifting Unseen-Scene Success from 9.7% to 43.3% — Shengguang Wu · 2026-11-02
- CUHK's CaPtain-4B Post-Trains Small VLMs as Code-as-Policy Robot Agents, Hitting 78.3 on LIBERO-PRO — CUOHK · 2026-11-02
- Foxconn deploys 10+ humanoid robots in its Texas AI server plants, built on NVIDIA GR00T — CyberRobooo · 2026-11-02
- Unitree robot dog guards home under a 'Beware of Dog' sign while the backup dog naps — yassineyousfi_ · 2026-11-02
- Probing 10 vision foundation models finds none supports a full spatial state — liuziwei7 · 2026-11-02
- JD's WorldBridge lifts robot task generalization to 70.7% with video plus sparse actions — Jun Cen · 2026-11-02