| 1 |
WoVR: World Models as Reliable Simulators for Post-Training VLA Policies with RL |
WoVR：利用世界模型作为可靠模拟器，通过强化学习后训练VLA策略 |
manipulation reinforcement learning imitation learning |
|
|
| 2 |
Rigidity-Based Multi-Finger Coordination for Precise In-Hand Manipulation of Force-Sensitive Objects |
提出基于刚性的多指协调框架，实现力敏感物体的精确灵巧操作 |
manipulation dexterous hand in-hand manipulation |
|
|
| 3 |
Direction Matters: Learning Force Direction Enables Sim-to-Real Contact-Rich Manipulation |
提出基于力方向学习的Sim-to-Real接触操作迁移框架，解决接触动力学差异问题。 |
manipulation sim-to-real |
✅ |
|
| 4 |
SemanticFeels: Semantic Labeling during In-Hand Manipulation |
SemanticFeels：结合视觉触觉，实现手内操作中的语义标签与形状重建 |
manipulation in-hand manipulation |
|
|
| 5 |
Learning Part-Aware Dense 3D Feature Field for Generalizable Articulated Object Manipulation |
提出Part-Aware 3D Feature Field (PA3FF)以提升可泛化的铰接物体操作能力 |
manipulation imitation learning diffusion policy |
|
|
| 6 |
A Latency-Aware Framework for Visuomotor Policy Learning on Industrial Robots |
提出一种延迟感知框架，用于工业机器人上的视觉伺服策略学习。 |
teleoperation policy learning multimodal |
|
|
| 7 |
ProAct: A Dual-System Framework for Proactive Embodied Social Agents |
ProAct：用于具身社交代理的主动式双系统框架 |
humanoid humanoid robot flow matching |
|
|
| 8 |
RoboAug: One Annotation to Hundreds of Scenes via Region-Contrastive Data Augmentation for Robotic Manipulation |
RoboAug：通过区域对比数据增强，仅需单张标注图像即可泛化至复杂机器人操作场景 |
manipulation |
✅ |
|