7 papers
Preserve Support, Not Correspondence: Dynamic Routing for Offline Reinforcement Learning
Zhancun Mu, Guangyu Zhao, Yiwu Zhong +1
One-step offline RL actors are attractive because they avoid backpropagating through long iterative samplers and keep inference cheap, but they still have to improve under a critic…
DeFlow: Decoupling Manifold Modeling and Value Maximization for Offline Policy Extraction
Zhancun Mu
We present DeFlow, a decoupled offline RL framework that leverages flow matching to faithfully capture complex behavior manifolds. Optimizing generative policies is computationally…
OpenHA: A Series of Open-Source Hierarchical Agentic Models in Minecraft
Zihao Wang, Muyao Li, Kaichen He +4
The choice of action spaces is a critical yet unresolved challenge in developing capable, end-to-end trainable agents. This paper first presents a large-scale, systematic compariso…
Scalable Multi-Task Reinforcement Learning for Generalizable Spatial Intelligence in Visuomotor Agents
Shaofei Cai, Zhancun Mu, Haiwen Xia +3
While Reinforcement Learning (RL) has achieved remarkable success in language modeling, its triumph hasn't yet fully translated to visuomotor agents. A primary challenge in RL mode…
ROCKET-2: Steering Visuomotor Policy via Cross-View Goal Alignment
Shaofei Cai, Zhancun Mu, Anji Liu +1
We aim to develop a goal specification method that is semantically clear, spatially sensitive, domain-agnostic, and intuitive for human users to guide agent interactions in 3D envi…
MineStudio: A Streamlined Package for Minecraft AI Agent Development
Shaofei Cai, Zhancun Mu, Kaichen He +4
Minecraft's complexity and diversity as an open world make it a perfect environment to test if agents can learn, adapt, and tackle a variety of unscripted tasks. However, the devel…