12 papers
SplineWAM: Adaptive Action Horizons for World Action Models via B-Spline Representations
Jun Guo, Xiaoshen Han, Qiwei Li +7
World action models (WAMs) are large embodied policies that jointly predict future video and the actions to execute, emitting a fixed-length action chunk per inference call. Such a…
Ordered Action Tokens for Visuomotor Policy Learning
Chaoqi Liu, Yue Zhao, Haonan Chen +4
Action tokenization maps continuous robot action chunks to discrete tokens and has become an important interface for modern visuomotor policies. Existing approaches either rely on…
B-spline Policy: Accelerating Manipulation Policies via B-spline Action Representations
Xiaoshen Han, Haoyu Xiong, Haonan Chen +4
In this work, we present B-spline Policy (BSP), an action representation designed for accelerating robot manipulation policies. Rather than predicting discrete-time action chunks,…
Structured 4D Latent Predictive Model for Robot Planning
Zhiyi Li, Peilin Wu, Xiaoshen Han +2
Video predictive models are emerging as a powerful paradigm in robotics, offering a promising path toward task generalization, long-horizon planning, and flexible decision-making.…
CoStream: Composing Simple Behaviors for Generalizable Complex Manipulation
Haonan Chen, Yuxiang Ma, Stephen Tian +7
Long-horizon, contact-rich complex manipulation tasks, such as seating a GPU into a PCIe slot, demand both millimeter high precision and out-of-the-box generalization to new tasks.…
OAT: Ordered Action Tokenization
Chaoqi Liu, Xiaoshen Han, Jiawei Gao +3
Autoregressive policies offer a compelling foundation for scalable robot learning by enabling discrete abstraction, token-level reasoning, and flexible inference. However, applying…
Enhancing Diffusion Policy with Classifier-Free Guidance for Temporal Robotic Tasks
Yuang Lu, Song Wang, Xiao Han +3
Temporal sequential tasks challenge humanoid robots, as existing Diffusion Policy (DP) and Action Chunking with Transformers (ACT) methods often lack temporal context, resulting in…
Manipulation as in Simulation: Enabling Accurate Geometry Perception in Robots
Minghuan Liu, Zhengbang Zhu, Xiaoshen Han +12
Modern robotic manipulation primarily relies on visual observations in a 2D color space for skill learning but suffers from poor generalization. In contrast, humans, living in a 3D…
DAWM: Diffusion Action World Models for Offline Reinforcement Learning via Action-Inferred Transitions
Zongyue Li, Xiao Han, Yusong Li +2
Diffusion-based world models have demonstrated strong capabilities in synthesizing realistic long-horizon trajectories for offline reinforcement learning (RL). However, many existi…
RoboGround: Robotic Manipulation with Grounded Vision-Language Priors
Haifeng Huang, Xinyi Chen, Yilun Chen +6
Recent advancements in robotic manipulation have highlighted the potential of intermediate representations for improving policy generalization. In this work, we explore grounding m…
ReSim: Generating High-Fidelity Simulation Data via 3D-Photorealistic Real-to-Sim for Robotic Manipulation
Xiaoshen Han, Junqiu Yu, Minghuan Liu +6
Real-world data collection for robotics is costly and resource-intensive, requiring skilled operators and expensive hardware. Simulations offer a scalable alternative but often fai…
DyDiff: Long-Horizon Rollout via Dynamics Diffusion for Offline Reinforcement Learning
Hanye Zhao, Xiaoshen Han, Zhengbang Zhu +4
With the great success of diffusion models (DMs) in generating realistic synthetic vision data, many researchers have investigated their potential in decision-making and control. M…