JD's WorldBridge lifts robot task generalization to 70.7% with video plus sparse actions
Jun Cen · hf · 2026-11-02
JD's Joy Future Academy introduces WorldBridge, a World Action Model combining separate video and action experts with a video-to-action bridge for robot task generalization. On RoboTwin 2.0, in-distribution video-action supervision alone yields 25.07% success on out-of-distribution tasks; adding videos of those tasks without action labels raises it to 55.90%; one action-labeled demo per task pushes it to 70.70%, showing broader video coverage enables transfer.
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
- CausalWAM Isolates Action Causes in World Action Models via Sparse Routing — causal-intelligence · 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