Score: 3

Training One Model to Master Cross-Level Agentic Actions via Reinforcement Learning

Published: December 10, 2025 | arXiv ID: 2512.09706v1

By: Kaichen He , Zihao Wang , Muyao Li and more

Potential Business Impact:

AI learns to use different tools for tasks.

Business Areas:
Machine Learning Artificial Intelligence, Data and Analytics, Software

The paradigm of agentic AI is shifting from engineered complex workflows to post-training native models. However, existing agents are typically confined to static, predefined action spaces--such as exclusively using APIs, GUI events, or robotic commands. This rigidity limits their adaptability in dynamic environments where the optimal granularity of interaction varies contextually. To bridge this gap, we propose CrossAgent, a unified agentic model that masters heterogeneous action spaces and autonomously selects the most effective interface for each step of a trajectory. We introduce a comprehensive training pipeline that integrates cold-start supervised fine-tuning with a Multi-Turn Group Relative Policy Optimization (GRPO) algorithm. This approach enables the agent to learn adaptive action switching--balancing high-level efficiency with low-level precision--without human-specified rules. Extensive experiments on over 800 tasks in the open-world Minecraft environment demonstrate that CrossAgent achieves state-of-the-art performance. By dynamically leveraging the strengths of diverse action spaces, our model significantly outperforms fixed-action baselines, exhibiting superior generalization and efficiency in long-horizon reasoning. All code and models are available at https://github.com/CraftJarvis/OpenHA

Country of Origin
πŸ‡ΈπŸ‡¬ πŸ‡¨πŸ‡³ Singapore, China

Repos / Data Links

Page Count
20 pages

Category
Computer Science:
Machine Learning (CS)