| 1 |
Stereo Multistage Spatial Attention for Real-Time Mobile Manipulation Under Visual Scale Variation and Disturbances |
提出基于立体多阶段空间注意力的深度预测学习方法，用于视觉尺度变化下的实时移动操作。 |
manipulation mobile manipulation imitation learning |
|
|
| 2 |
MSACT: Multistage Spatial Alignment for Stable Low-Latency Fine Manipulation |
MSACT：多阶段空间对齐实现低延迟稳定精细操作 |
manipulation bi-manual bimanual manipulation |
|
|
| 3 |
Learning while Deploying: Fleet-Scale Reinforcement Learning for Generalist Robot Policies |
提出LWD框架，用于通用机器人策略在部署中持续学习，提升真实环境下的泛化能力。 |
manipulation dual-arm reinforcement learning |
|
|
| 4 |
Embodied Interpretability: Linking Causal Understanding to Generalization in Vision-Language-Action Models |
提出干预显著性得分(ISS)和干扰质量比(NMR)，诊断VLA模型中的因果错位问题。 |
manipulation vision-language-action VLA |
|
|
| 5 |
PrefMoE: Robust Preference Modeling with Mixture-of-Experts Reward Learning |
提出PrefMoE，通过混合专家模型提升偏好学习在噪声数据下的鲁棒性 |
locomotion manipulation reinforcement learning |
|
|
| 6 |
Recovering Hidden Reward in Diffusion-Based Policies |
EnergyFlow：基于扩散模型的策略学习框架，实现隐式奖励恢复 |
manipulation reinforcement learning inverse reinforcement learning |
✅ |
|
| 7 |
A Model-based Visual Contact Localization and Force Sensing System for Compliant Robotic Grippers |
提出一种基于模型的视觉接触定位和力感应系统，用于柔性机器人夹爪 |
manipulation 3D reconstruction |
|
|