2026-06-19原文2 min readUpdated: 2026-06-19

Playful Agentic Robot Learning

This paper introduces Playful Agentic Robot Learning, where an embodied coding agent uses self-directed play as a continual skill-learning stage. The proposed RATs framework proposes exploratory tasks, executes code policies, verifies progress, and distills successful executions into a persistent skill library. Experiments show significant improvements on LIBERO-PRO and MolmoSpaces, and transferable skills boost performance without fine-tuning.

SourcearXiv RoboticsAuthor: Junyi Zhang, Jiaxin Ge, Hanjun Yoo, Letian Fu, Zihan Yang, Yaowei Liu, Raj Saravanan, Shaofeng Yin, Justin Yu, Dantong Niu, Zirui Wang, Roei Herzig, Ken Goldberg, Yutong Bai, David M. Chan, Ion Stoica, Angjoo Kanazawa, Jiahui Lei, Haiwen Feng, Trevor Darrell

[2606.19419] Playful Agentic Robot Learning

[Submitted on 17 Jun 2026]

Title:Playful Agentic Robot Learning

View a PDF of the paper titled Playful Agentic Robot Learning, by Junyi Zhang and 19 other authors

View PDF HTML (experimental)

Abstract:Current agentic robot systems can write executable Code-as-Policy programs, observe feedback, and revise behavior across multiple attempts, but they remain largely task-driven: reusable skills are acquired only after explicit instructions. We study Playful Agentic Robot Learning, where an embodied coding agent uses self-directed play as a continual skill-learning stage before downstream tasks arrive. We introduce RATs, Robotics Agent Teams designed for play-time skill acquisition. During play, RATs proposes novel yet learnable exploratory tasks, plans and executes robot-code policies, verifies intermediate progress, diagnoses failures, retries with dense, step-level feedback, and distills successful executions into a persistent code skill library. At test time, the agent reuses relevant skills from this frozen library to help solve new tasks. Experiments in LIBERO-PRO and MolmoSpaces show that play-learned skills improve held-out downstream tasks over no-play and random-play baselines, with 20.6 and 17.0 percentage-point gains over CaP-Agent0 on LIBERO-PRO and MolmoSpaces, respectively. Moreover, the learned skills can be plugged into other inference-time Code-as-Policy agents by simply retrieving them into the context, improving RoboSuite and real-world transfer by 8.9 and 8.8 points, respectively, without finetuning the underlying model.

Comments: Project page: this https URL

Subjects:

Robotics (cs.RO); Artificial Intelligence (cs.AI)

Cite as: arXiv:2606.19419 [cs.RO]

(or arXiv:2606.19419v1 [cs.RO] for this version)

https://doi.org/10.48550/arXiv.2606.19419

arXiv-issued DOI via DataCite (pending registration)

Submission history

From: Junyi Zhang [view email] [v1] Wed, 17 Jun 2026 17:55:23 UTC (5,893 KB)

Full-text links:

Access Paper:

View a PDF of the paper titled Playful Agentic Robot Learning, by Junyi Zhang and 19 other authors

View PDF

HTML (experimental)

TeX Source

view license

Current browse context:

cs.RO

new | recent | 2026-06

Change to browse by:

cs cs.AI

References & Citations

NASA ADS

Google Scholar

Semantic Scholar

Data provided by:

Bibliographic Tools

Bibliographic and Citation Tools

Bibliographic Explorer Toggle

Bibliographic Explorer (What is the Explorer?)

Connected Papers Toggle

Connected Papers (What is Connected Papers?)

Litmaps Toggle

Litmaps (What is Litmaps?)

scite.ai Toggle

scite Smart Citations (What are Smart Citations?)

Code, Data, Media

Code, Data and Media Associated with this Article

alphaXiv Toggle

alphaXiv (What is alphaXiv?)

Links to Code Toggle

CatalyzeX Code Finder for Papers (What is CatalyzeX?)

DagsHub Toggle

DagsHub (What is DagsHub?)

GotitPub Toggle

Gotit.pub (What is GotitPub?)

Huggingface Toggle

Hugging Face (What is Huggingface?)

ScienceCast Toggle

ScienceCast (What is ScienceCast?)

Demos

Replicate Toggle

Replicate (What is Replicate?)

Spaces Toggle

Hugging Face Spaces (What is Spaces?)

Spaces Toggle

TXYZ.AI (What is TXYZ.AI?)