| 1 |
ConRFT: A Reinforced Fine-tuning Method for VLA Models via Consistency Policy |
提出ConRFT,通过强化微调提升VLA模型在复杂操作任务中的鲁棒性。 |
manipulation reinforcement learning behavior cloning |
✅ |
|
| 2 |
HAMSTER: Hierarchical Action Models For Open-World Robot Manipulation |
HAMSTER:用于开放世界机器人操作的分层动作模型 |
manipulation vision-language-action VLA |
|
|
| 3 |
Temporal Representation Alignment: Successor Features Enable Emergent Compositionality in Robot Instruction Following |
利用后继特征和时间对齐损失,实现机器人指令跟随中的涌现组合性 |
manipulation reinforcement learning instruction following |
|
|
| 4 |
Motion Planning of Nonholonomic Cooperative Mobile Manipulators |
提出一种非完整约束协同移动机械臂的实时运动规划技术,用于复杂环境下的物体运输。 |
locomotion manipulation model predictive control |
|
|
| 5 |
Generating Physically Realistic and Directable Human Motions from Multi-Modal Inputs |
提出Masked Humanoid Controller (MHC),从多模态输入生成逼真可控的人体运动 |
humanoid humanoid control imitation learning |
|
|
| 6 |
Implicit Physics-aware Policy for Dynamic Manipulation of Rigid Objects via Soft Body Tools |
提出基于隐式物理感知的策略,用于软体工具动态操作刚性物体 |
manipulation |
|
|
| 7 |
Towards Learning Scalable Agile Dynamic Motion Planning for Robosoccer Teams with Policy Optimization |
提出基于策略优化的可扩展敏捷动态运动规划方法,用于机器人足球队 |
motion planning |
|
|