Stanford's RoboArchitect Auto-Designs Robot Harnesses, Lifting Unseen-Scene Success from 9.7% to 43.3%
Shengguang Wu · hf · 2026-11-02
Stanford researchers present RoboArchitect, which uses a coding model to automatically design robot agent harnesses—the control code deciding when to query a frozen policy, consult vision tools, or command the arm directly.
- Approach: The coding model observes rollouts, diagnoses failures, and iteratively revises the harness; the harness evolves on a set of episodes and is evaluated on held-out scenes and tasks.
- Results: Without updating any weights, average success on unseen scenes rises from 9.7% to 43.3% (4.5×) and on unseen tasks from 12.7% to 40.2% (3.2×), across multiple benchmarks and policies.
- Highlight: Gains are largest far from the policy's training data; the evolved harness around a general-purpose policy often outperforms even the finetuned version, and intuitive control strategies emerge automatically.
More from Embodied
- 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
- JD's WorldBridge lifts robot task generalization to 70.7% with video plus sparse actions — Jun Cen · 2026-11-02