Embodied AI

Abstract: In this paper, we introduce an innovative vector quantization based action tokenizer built upon the largest-scale action trajectory dataset to date, leveraging over 100 times more data than previous approaches.

Abstract: World models serve as essential building blocks toward Artificial General Intelligence (AGI), enabling intelligent agents to predict future states and plan actions by simulating complex physical interactions. However, existing interactive models primarily predict visual observations, thereby neglecting crucial hidden states like geometric structures and spatial coherence.

Abstract: Learning latent motion from Internet videos is crucial for building generalist robots. However, existing discrete latent action methods suffer from information loss and struggle with complex and fine-grained dynamics. We propose CoMo, which aims to learn more informative continuous motion representations from diverse, internet-scale videos.

A geometric-aware unified world model, capable of 4D reconstruction, action-conditioned prediction, and visual planning.

Abstract: Learning from multiple domains is a primary factor that influences the generalization of a single unified robot system. In this paper, we aim to learn the trajectory prediction model by using broad out-of-domain data to improve its performance and generalization ability.

A novel representation learning framework that emphasizes the importance of 3D spatial awareness in embodied AI.

Extensive experiments prove that point cloud observations are beneficial for robot learning.

Abstract: A key challenge in robotic manipulation in open domains is how to acquire diverse and generalizable skills for robots. Recent research in one-shot imitation learning has shown promise in transferring trained policies to new tasks based on demonstrations.

Building open-ended agents with internet-scale knowledge in Minecraft.