# Learning and Transferring Closed-Loop Robot Software

arXiv: https://arxiv.org/abs/2609.19906 · HTML: https://mlsanity.com/paper/2609.19906/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 22 of its 25 references that are in the map.
- First posted on arXiv 17 Sep 2026.

## Builds on (most cited first, 22 of 22)

- [SAM 2: Segment Anything in Images and Videos](https://mlsanity.com/paper/2408.00714/) (2024)
- [Diffusion Policy: Visuomotor Policy Learning via Action Diffusion](https://mlsanity.com/paper/2303.04137/) (2023)
- [π0: A Vision-Language-Action Flow Model for General Robot Control](https://mlsanity.com/paper/2410.24164/) (2024)
- [π0.5: a Vision-Language-Action Model with Open-World Generalization](https://mlsanity.com/paper/2504.16054/) (2025)
- [Depth Anything V2](https://mlsanity.com/paper/2406.09414/) (2024)
- [GR00T N1: An Open Foundation Model for Generalist Humanoid Robots](https://mlsanity.com/paper/2503.14734/) (2025)
- [Code as Policies: Language Model Programs for Embodied Control](https://mlsanity.com/paper/2209.07753/) (2022)
- [VoxPoser: Composable 3D Value Maps for Robotic Manipulation with Language Models](https://mlsanity.com/paper/2307.05973/) (2023)
- [RoboCasa: Large-Scale Simulation of Everyday Tasks for Generalist Robots](https://mlsanity.com/paper/2406.02523/) (2024)
- [ReKep: Spatio-Temporal Reasoning of Relational Keypoint Constraints for Robotic Manipulation](https://mlsanity.com/paper/2409.01652/) (2024)
- [CaP-X: A Framework for Benchmarking and Improving Coding Agents for Robot Manipulation](https://mlsanity.com/paper/2603.22435/) (2026)
- [RoboCasa365: A Large-Scale Simulation Framework for Training and Benchmarking Generalist Robots](https://mlsanity.com/paper/2603.04356/) (2026)
- [ASPIRE: Agentic /Skills Discovery for Robotics](https://mlsanity.com/paper/2607.00272/) (2026)
- [Demo2Code: From Summarizing Demonstrations to Synthesizing Code via Extended Chain-of-Thought](https://mlsanity.com/paper/2305.16744/) (2023)
- [Lifelong Robot Library Learning: Bootstrapping Composable and Generalizable Skills for Embodied Control with Language Models](https://mlsanity.com/paper/2406.18746/) (2024)
- [Playful Agentic Robot Learning](https://mlsanity.com/paper/2606.19419/) (2026)
- [RHO: Your Coding Agent is Secretly a Roboticist](https://mlsanity.com/paper/2606.16458/) (2026)
- [Act-Observe-Rewrite: Multimodal Coding Agents as In-Context Policy Learners for Robot Manipulation](https://mlsanity.com/paper/2603.04466/) (2026)
- [Memory Transfer Planning: LLM-driven Context-Aware Code Adaptation for Robot Manipulation](https://mlsanity.com/paper/2509.24160/) (2025)
- [Iterative Program Synthesis for Adaptable Social Navigation](https://mlsanity.com/paper/2103.04880/) (2021)
- [Programmatic Imitation Learning from Unlabeled and Noisy Demonstrations](https://mlsanity.com/paper/2303.01440/) (2023)
- [A Few Words Go a Long Way: Language Guided Robot Policy Synthesis](https://mlsanity.com/paper/2607.23784/) (2026)

## Built on by (most cited first, 2 of 2)

- [In-Context Learning for Robots: Methods and Applications](https://mlsanity.com/paper/2609.36012/) (2026)
- [SEAR: Self-Evolving Agentic Robotics](https://mlsanity.com/paper/2610.18792/) (2026)

## Abstract (arXiv)

Closed-loop robot policies require observation processing, state management, and situation-dependent branching, making them costly to design and tune manually. Although coding agents increasingly support control-code generation and optimization, it remains unclear whether implementations improved on source tasks also support policy acquisition for new tasks. We study this question by treating complete closed-loop implementations as reusable execution experience. For each source task, a coding agent generates policy code from a few successful demonstrations and iteratively improves it using simulation feedback. The validation-selected implementations are retained in a software archive. For new tasks, the agent generates and improves policies using archived implementations, target demonstrations, and execution feedback. The resulting policy is then frozen and executes without further model calls. Across four source tasks in RoboCasa, iterative optimization increases mean success from 28.3% to 64.2%. Across nine target tasks and three independent runs, mean success is 45.2% without references, 41.5% with initial source code, and 57.0% with optimized source code. Optimized references outperform initial references in all three runs on the nine-task average, with a mean gain of 15.6 percentage points. These results demonstrate the value of execution-improved software as a resource for acquiring new policies in this setting, although initial references remain better on two target tasks when averaged across runs.