The paper introduces Immiscible Diffusion Policy, a label‑free training add‑on that prevents diffusion policies from collapsing to a single action modality. By assigning action‑noise pairs, it preserves distinct noise‑to‑action routes without altering the policy architecture or inference. Experiments on five simulated and two real‑world humanoid manipulation tasks show significant improvements in maintaining multimodal action distributions, increasing non‑dominant modality proportions by up to 14.6× and recovering missing modalities in four‑modality tasks.
By Xiao Zhang, Yuxin Chen, Zhixuan Liang, Guojian Zhan, Chenran Li, Chenfeng Xu, Masayoshi Tomizuka, Yiheng Li
arXiv:2512. 07212v3 Announce Type: replace Abstract: Imitation learning with diffusion models has advanced robotic control by capturing the multi-modal action distributions.
By Zhaoyang Liu, Mokai Pan, Zhongyi Wang, Kaizhen Zhu, Haotao Lu, Haipeng Zhang, Jingya Wang, Ye Shi
arXiv:2607. 17257v1 Announce Type: cross Abstract: Diffusion policies have shown strong potential for robotic imitation learning, and recent extensions incorporate additional modalities to improve manipulation performance.
By Zihao He, Hongjie Fang, Shirun Tang, Cewu Lu, Haoshu Fang
We propose Ambient Diffusion Policy, a simple and principled method for imitation learning from suboptimal data in robotics. High-quality, task-specific robot data is expensive and time-consuming to collect, while suboptimal datasets with lower-quality or out-of-distribution demonstrations are abundant.
arXiv:2606. 12365v1 Announce Type: cross Abstract: We propose Ambient Diffusion Policy, a simple and principled method for imitation learning from suboptimal data in robotics.
By Adam Wei, Nicholas Pfaff, Thomas Cohn, Arif Kerem Day{\i}, Constantinos Daskalakis, Giannis Daras, Russ Tedrake
arXiv:2606. 00336v1 Announce Type: new Abstract: We propose Parameterized Diffusion Policy (PDP), a framework for learning diffusion policies conditioned on low-dimensional, continuous parameters embedded in a learned behavior manifold.
By Renhao Zhang, Haotian Fu, Mingxi Jia, George Konidaris, Yilun Du, Bruno Castro da Silva
arXiv:2606. 16447v1 Announce Type: cross Abstract: Imitation learning has enabled highly-dexterous robotic manipulation from RGB observations.
By Abhinav Agarwal, Adam Wei, Taylan Kargin, Michael Zeng, Cole Becker, Arif Kerem Dayi, Pablo Parrilo, Asuman Ozdaglar, Russ Tedrake
arXiv:2606. 08657v1 Announce Type: cross Abstract: Diffusion-based visuomotor policies operating directly in raw action spaces conflate scene comprehension with trajectory generation within a single denoising process.
By Zhexuan Zhou, Yichen Lai, Jinhao Zhang, Huizhe Li, Youmin Gong, Jie Mei
arXiv:2506. 20668v3 Announce Type: replace-cross Abstract: We propose DemoDiffusion, a simple method for enabling robots to perform manipulation tasks by imitating a single human demonstration, without requiring task-specific training or paired human-robot data.
By Sungjae Park, Homanga Bharadhwaj, Shubham Tulsiani
arXiv:2602. 09580v4 Announce Type: replace-cross Abstract: Real-world fine-tuning of dexterous manipulation policies remains challenging due to limited real-world interaction budgets and highly multimodal action distributions.
By Chenyu Yang, Denis Tarasov, Davide Liconti, Romain Guntz, Hehui Zheng, Robert K. Katzschmann
arXiv:2605. 12236v2 Announce Type: replace-cross Abstract: Fine-tuning pre-trained robot policies with reinforcement learning (RL) often inherits the bottlenecks introduced by pre-training with behavioral cloning (BC), which produces narrow action distributions that lack the coverage necessary for downstream exploration.
By Matthew M. Hong, Jesse Zhang, Anusha Nagabandi, Abhishek Gupta
arXiv:2607. 19919v1 Announce Type: cross Abstract: We propose Diffusion ReRoll, a diffusion-based framework for robotic sequential prediction that enables revisable denoising over horizons.
By Seonsoo Kim, Seongil Hong, Jun-Gill Kang