This paper presents ShapeLLM, the first 3D Multimodal Large Language Model (LLM)
designed for embodied interaction, exploring a universal 3D object understanding with 3D …

Robot manipulation relies on accurately predicting contact points and end-effector directions
to ensure successful operation. However learning-based robot manipulation trained on a …

We present ChainedDiffuser, a policy architecture that unifies action keypose prediction and
trajectory diffusion generation for learning robot manipulation from demonstrations. Our …

Abstract We propose Neural 3D Articulated object Prior (NAP), the first 3D deep generative
model to synthesize 3D articulated object models. Despite the extensive research on …

Enabling robotic manipulation that generalizes to out-of-distribution scenes is a crucial step
toward open-world embodied intelligence. For human beings, this ability is rooted in the …

Understanding the continuous states of objects is essential for task learning and planning in
the real world. However, most existing task learning benchmarks assume discrete (eg …

This work proposes a retrieve-and-transfer framework for zero-shot robotic manipulation,
dubbed RAM, featuring generalizability across various objects, environments, and …

Learning a universal manipulation policy encompassing doors with diverse categories,
geometries and mechanisms, is crucial for future embodied agents to effectively work in …

Affordance detection presents intricate challenges and has a wide range of robotic
applications. Previous works have faced limitations such as the complexities of 3D object …

Garment manipulation (eg unfolding folding and hanging clothes) is essential for future
robots to accomplish home-assistant tasks while highly challenging due to the diversity of …