Robot Learning

VQ-VLA: Improving Vision-Language-Action Models via Scaling Vector-Quantized Action Tokenizers
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.
Yating Wang, Haoyi Zhu 朱皓怡, Mingyu Liu, Jiange Yang, Hao-Shu Fang, Tong He
Cite Website GitHub PDF arXiv
VQ-VLA: Improving Vision-Language-Action Models via Scaling Vector-Quantized Action Tokenizers
CoMo: Learning Continuous Latent Motion from Internet Videos for Scalable Robot Learning
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.
Jiange Yang, Yansong Shi, Haoyi Zhu 朱皓怡, Mingyu Liu, Kaijing Ma, Yating Wang, Gangshan Wu, Tong He, Limin Wang
Cite PDF arXiv
CoMo: Learning Continuous Latent Motion from Internet Videos for Scalable Robot Learning
Tra-MoE: Learning Trajectory Prediction Model from Multiple Domains for Adaptive Policy Conditioning
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.
Jiange Yang, Haoyi Zhu 朱皓怡, Yating Wang, Gangshan Wu, Tong He, Limin Wang
Cite PDF Arxiv
Tra-MoE: Learning Trajectory Prediction Model from Multiple Domains for Adaptive Policy Conditioning
Point Cloud Matters: Rethinking the Impact of Different Observation Spaces on Robot Learning
Extensive experiments prove that point cloud observations are beneficial for robot learning.
Haoyi Zhu 朱皓怡, Yating Wang, Di Huang, Weicai Ye, Wanli Ouyang, Tong He
Cite Project Github PDF Arxiv
Point Cloud Matters: Rethinking the Impact of Different Observation Spaces on Robot Learning
RH20T: A Robotic Dataset for Learning Diverse Skills in One-Shot
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.
Hao-Shu Fang, Hongjie Fang, Zhenyu Tang, Jirong Liu, Junbo Wang, Haoyi Zhu 朱皓怡, Cewu Lu
Cite Project PDF Arxiv
RH20T: A Robotic Dataset for Learning Diverse Skills in One-Shot
14,968 Total Pageviews