Paper page - AgenticRoboRL: Bringing Coding-Agent Post-Training to Agentic Robotics

Source: https://huggingface.co/papers/2610.28589

Abstract

AgenticRoboRL improves robot planners via supervised fine-tuning and multi-turn reinforcement learning across diverse harnesses, significantly boosting success rates without external memory.

Agentic roboticssystems combine planners withrobot harnessesto perform complex manipulation tasks. Recent work improves these systems through better harnesses and external memory, but relies on powerful closed-source models as planners and does not improve the planners’ own capabilities. In contrast, coding agents have improved substantially by post-training the model itself. We therefore investigate whether post-training can also improve robot planners. However, applying it acrossrobot harnessesrequires accommodating different execution mechanisms, includingVLA skill invocationandcontrol-code generation. To this end, we introduceAgenticRoboRL, a framework for post-training robot planners across different harnesses. The training recipe combinessupervised fine-tuningwithmulti-turn reinforcement learning. To accommodate different execution mechanisms,harness-specific adaptersconnect recorded planner interactions with task outcomes, enabling a common RL training pipeline while keeping execution components fixed. OnLIBERO-PRO,AgenticRoboRLraises the success rate of a 9B-parameter planner from 18.1% to 71.0% without external memory, approaching the 72.1% achieved by Codex (GPT-5.5) with external memory. These results demonstrate post-training as an effective approach for improving robot planners. We will release the framework and trained planner checkpoints.

View arXiv pageView PDFProject pageGitHub11Add to collection

Get this paper in your agent:

hf papers read 2610\.28589

Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash

Models citing this paper1

#### ciciciciliu/AgenticRoboRL-Qwen3.5-9B Updatedabout 22 hours ago • 3

Datasets citing this paper0

No dataset linking this paper

Cite arxiv.org/abs/2610.28589 in a dataset README.md to link it from this page.

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2610.28589 in a Space README.md to link it from this page.

Collections including this paper0

No Collection including this paper

Add this paper to acollectionto link it from this page.