AgenticRoboRL enhances the performance of robotic planners
The story so far
qwen's AgenticRoboRL framework enhances robot planners through supervised fine-tuning and multi-round reinforcement learning. On LIBERO-PRO, it increased the success rate of a 9B parameter planner from 18.1% to 71.0%, approaching Codex (GPT-5.5) using external memory achieved 72.1%. This framework does not rely on external memory, adapts to different execution mechanisms, and achieves a general reinforcement learning training process by recording planner interactions and connecting task results.
AI-written from the reports · Updated 27 days ago
Coverage timeline
Follow the coverage to see every side of the story.
- Hugging Face Daily PapersAgenticRoboRL: Bringing Coding-Agent Post-Training to Agentic Robotics
AgenticRoboRL framework improves robot planner performance via supervised fine-tuning and multi-round reinforcement learning, boosting the success rate of a 9B parameter planner from 18.1% to 71.0% on LIBERO-PRO without external memory, approaching Codex (GPT-5.5)'s 72.1% with external memory. The framework adapts to different execution mechanisms, linking recorded planner interactions with task outcomes to achieve a universal reinforcement learning training process through adapters.