RMBench is introduced, a simulation benchmark comprising 9 manipulation tasks that span multiple levels of memory complexity, enabling systematic evaluation of policy memory capabilities and Mem-0, a modular manipulation policy with explicit memory components designed to support controlled ablation studies.
Tian-Xing Chen, Yu-Ran Wang, Ming-Lei Li et al.· arXiv.org· 62 citations· ⚡23
M-Diffusion Planner is proposed with a dual-decoupling design: separating intent understanding from trajectory planning, and general planning capability learning from strategy alignment, to mitigate degradation of pretrained planning capabilities.
Experiments show that current VLA policies struggle with zero-shot personalization, while SFT makes preferences learnable, and a controlled formulation and taxonomy of observable preferences in simulation is proposed, covering motion, skill, and user-profile choices.
Zhi-Xu Li, Ke Tang, Li-Tian Gong et al.· 0 citations
T2-Z is introduced for test-time optimization of environment latents, which augments a robotic world model with a learnable environment latent to condition world predictions, and can efficiently infer environment latents from as little as a single test-time interaction trajectory.
Chen-Yuan Zhou, Yi-Ning Hong, Huang Huang et al.· 0 citations
Reach audiences
Advertise in front of researchers, engineers, and readers.
This work introduces CoDA, a framework in which a simulation data engine and an action policy co-evolve through a shared heterogeneous memory of environment construction, policy failures, and learning progress, and proposes Text2potential to convert these diagnoses into staged potentials for action-chunk value learning...
Tianjiao Yu, Zhi-Lin Ou, Yi-Fan Shen et al.· 0 citations
Recent 3D world models increasingly use features from geometry foundation models (GFMs) as predictive world states, benefiting from their rich geometric priors. However, the view dependence of these features entangles scene dynamics with camera-induced variation, making the resulting state evolution harder to predict....
Hongchang Liu, Jian Liu, Yujuan Ding et al.· 0 citations
How can robots autonomously collect the contact-rich experience needed to learn dexterous manipulation? We present DSP: Dexterity from supervised play, a framework in which a robot learns a play policy from minutes of human play through behavior cloning, then deploys it autonomously to collect hours of interaction data...
Botao He, Yiming Dou, Yiannis Aloimonos et al.· 0 citations
An informed prior is built that targets preference queries at the features that need resolving, and each query is accompanied with an explanation that focuses the human's attention on exactly those features.
Helena A. Merker, Nick Walker, Andreea Bobu· 0 citations
Current humanoid dexterous manipulation remains largely confined to end-effectors and simple grasps, falling short of humans'ability to engage the entire body by bracing and sliding bulky objects against the arms and torso. We introduce Whole-Body Dexterity (WBD), a general framework that uses the robot's body and appe...
Conor Mc Gartoll, An Dang, Pranay Thangeda et al.· 0 citations
Learning from demonstration is an effective strategy for endowing robots with manipulation capabilities but requires training on datasets comprised of massive amounts of demonstrations. Existing demonstration collection approaches either require constant human involvement, for example, via teleoperation or are speciali...
Scaling robot manipulation models requires large amounts of interaction data, yet collecting diverse real-robot experience remains costly. Human manipulation videos, in contrast, provide abundant and diverse task experience that could potentially benefit robot learning. However, differences in morphology, action spaces...
Xingyu Zhang, Yingchao Li, Peizheng Guo et al.· 0 citations
Indoor service robots often enter unmapped buildings whose layouts and task constraints are already familiar to users. We address this human-familiar, robot-unseen setting with AVALON, a topological navigation framework organized around human-seeded layered spatial memory, node-verified topological execution, and task-...
Jing Xia, Changyan Li, Yuduo Wang et al.· 0 citations
A new system identifies and corrects unreliable object descriptions stored in a robot’s memory, improving the robot’s ability to interact with its environment.
Robots are getting smarter, but how can their hardware match that growth? New Microsoft Research findings show that moving AI inference beyond the robot can improve task success, boost efficiency, and support more advanced physical AI workloads. The post Offloaded inference for real-world physical AI robotics appeared first on Microsoft Research.
