Code as World
Code as World organizes interaction with the physical environment into two levels of code. L1 provides the basic operations for observing and acting; L2 combines these operations into executable Python procedures. Programmatic control sits within the broader set of embodied control paradigms.
The environment exposes four primitives for reading the task, observing the scene, executing an action, and resetting the environment. These operations form the fixed interface available to the agent. Reset is restricted to Playground, where it enables another attempt on the same scene.
In RoboDojo, observations contain RGB images and robot state, including joint states, end-effector poses, and gripper command states. Actions specify joint or end-effector targets together with gripper commands; the native controller handles inverse kinematics and interpolation. Object ground-truth poses, depth, and camera calibration are unavailable through this interface.
Deep Dive into Astra’s Failures
To understand where coding agents struggle in physical interaction, we evaluate GPT-6-Astra with xhigh reasoning effort in the code workspace. The evaluation is on RoboDojo-sim tasks spanning Generalization, Precision, Long-Horizon, Memory, and Open, with 5 scenes per task. Astra receives the task description and robot-interface documentation, without the task wiki. Each episode starts with a fresh agent, workspace, and simulator, permits up to 100 accepted code executions within the task’s native action allowance, and disables reset.
Astra achieves an average success rate of 8.50%. We examine its submitted programs, camera observations, and feedback to locate where plausible plans fail during physical execution. Two illustrative episodes expose different gaps between issuing a valid command and advancing the task.
In tic-tac-toe, Astra attempts to pick up a ring and place it on the board. The gripper reaches its target positions, yet two successive attempts leave the ring on the table (actions 216 and 283). After inspecting a wrist-camera view and adjusting the grasp, Astra first visibly retains the ring at action 367, consuming one-third of its 1,100-action budget. The episode later times out without completing the board. This trace shows how constructing a local manipulation routine can compete with task completion. It motivates retaining a parameterized procedure that combines approach, alignment, closure, and verification of object pickup, so that subsequent attempts can build on a working operation.
Self-Improvement Through Harness Evolution
One Playground scene hosts a persistent loop of code editing, execution, and feedback. Robot state and Python definitions persist between submissions; reset allows another attempt on the same scene. The agent updates Skills and Lessons and applies them in subsequent attempts. A successful operation can contribute a parameterized procedure; an unsuccessful attempt can reveal a useful condition or suggest a recovery strategy. Even an unfinished task can yield useful findings; tentative explanations remain hypotheses.
At the end of Playground, only Skills and Lessons are carried into Test as read-only artifacts. On the right, each unseen scene of the same task starts with a fresh agent, workspace, Python namespace, and simulator. The agent uses the frozen harness to construct code and adapts that code through execution feedback. Reset is disabled, and Test episodes do not update the harness or share new experience. This separates within-scene adaptation from the contribution of knowledge acquired during Playground.
The same frozen artifacts can be supplied to another executor because their contents are explicit procedures and guidance. Cross-model reuse requires no additional Playground exploration, parameter updates, or learned adapter. We compare self-generated and transferred harnesses under the same executor and Test conditions to assess whether the acquired knowledge remains useful across models.
Experiments
We evaluate EmbodiedEvo on RoboDojo-Sim to address the following research questions:
- How well does EmbodiedEvo perform compared with VLA- and WAM-based methods?
- Which capabilities improve through harness evolution, and which tasks remain difficult?
- Can acquired experience transfer across models and tasks?
Experimental Setup
RoboDojo-Sim is a challenging benchmark for evaluating generalist robot manipulation policies. It comprises 12 Generalization, 8 Precision, 8 Long-Horizon, 6 Memory, and 8 Open tasks. Each harness (Skills + Lessons) is acquired on one Playground scene layout distinct from the Test layouts.
Executors are GPT-6-Astra at xhigh and medium effort, GPT-5.6-sol at xhigh, and GPT-5.6-luna at xhigh, abbreviated Astra, Sol, and Luna.
Discussion
Code exposes the parameters, checks, and control flow through which a physical strategy is executed. This gives feedback a concrete target for revision: after a dropped load, the stacking agent retains bounded motion and checks object retention during later execution. Parameterized procedures can then preserve a useful operation while allowing its targets to change with the scene. Lessons complement this representation by recording when to apply or reconsider it. Raw History achieves similar observed success, while Skills/Lessons reduce total Test token use by 11.3%. This suggests that organizing interaction into accessible procedures and guidance is a practical benefit of the representation; code alone does not guarantee valid physical assumptions.
Conclusion
We presented EmbodiedEvo, a framework for coding agents to turn physical interaction into reusable Skills and Lessons without updating model parameters. On RoboDojo-Sim, EmbodiedEvo achieves 44.48 Average Score and 37.82% SR. Paired evaluations show consistent gains across different input settings. Skills and Lessons achieve comparable observed performance to raw interaction history at 53.0% lower Test execution cost. These findings suggest that harness evolution improves the starting point of task solving by preserving executable operations and the conditions for applying them. Precise contact and reliable long-horizon execution remain important directions for extending this approach.
Capability Changes
Execution records underpin the capability analysis. All cases use Astra/xhigh; paired comparisons use the same scene. The evidence bundle retains the original programs, feedback, file reads, and camera frames. These selected cases illustrate mechanisms, not failure frequencies or isolated causal effects.
Experience Reuse and Its Limits
Successful reuse, cross-model transfer, and a regression are compared. The complete retained grasp function and two Lessons used in the successful stacking trace are also provided.
The frozen lesson records, “The arm reached its target, but load retention failed.” Test execution 2 includes ee_move.py and pick_from_above.py verbatim. Execution 9 uses an 8 mm translation bound after visual retention checking; execution 15 completes the stack. This establishes actual reuse, without isolating the bound’s causal contribution.
Frequently Asked Questions
What are Skills and Lessons in EmbodiedEvo? Skills are reusable parameterized procedures, while Lessons record conditions for applying or reconsidering them.
What information does the RoboDojo interface provide? It provides RGB images and robot state, including joint states, end-effector poses, and gripper command states. It does not provide object ground-truth poses, depth, or camera calibration.
How does EmbodiedEvo separate learning from testing? Skills and Lessons are acquired in Playground and frozen as read-only artifacts for Test. Test scenes use fresh agents, workspaces, Python namespaces, and simulators, with reset disabled.
What limitations remain for EmbodiedEvo? Precise contact and reliable long-horizon execution remain important directions for extending the approach.