Survey of Robot Learning: Weights vs. Self-Writing Code Skills
Gaytri Jena · hf · 2026-08-08
This academic survey organizes the field of robot learning around two main bets: baking competence into frozen weights (VLA models) versus agents that write and refine their own executable skills as code.
Arranging code-as-policy methods by their degree of self-improvement, the paper explores everything from zero-shot program synthesis to open-ended loops combining execution feedback and evolutionary search. It also highlights open challenges in the emerging skill economy, such as cross-embodiment portability, safety verification, and standardization.
More from Embodied
- Unitree Lists Humanoid Robot H2 for $29,900 — Grady_Booch · 2026-08-08
- Robot Shows Impressive Force Control by Handling Fragile Fortune Cookie — JasonMa2020 · 2026-08-08
- Eldercare is the best deployment area for robotics right now — carlosdponx · 2026-08-08
- Science Corp's Retinal Implant Helps Blind Read Novels; CEO Max Hodak Shares Startup Playbook — ycombinator · 2026-08-08
- Legs vs Wheels: Which Makes More Sense for Home Robots? — chris_j_paxton · 2026-08-08
- Open Source WebXR Project Turns Phones and VR Headsets into Robot Arm Teleoperation Devices — tom_doerr · 2026-08-08