Fast dexterous grasping while a mobile base remains in motion requires coordinated whole-body control and rapid adaptation to physical contact and this work proposes FastGrasp, a two-stage learning framework that integrates grasp guidance, whole-body control, and tactile feedback.
Heng Tao, Yi-Ming Zhong, Zemin Yang et al.· arXiv.org· 0 citations
Vision-Language-Action models (VLAs) represent a significant frontier in embodied intelligence, aiming to bridge digital knowledge with physical-world interaction. Despite their remarkable performance, foundational VLAs are hindered by the prohibitive computational and data demands inherent to their large-scale archite...
Zhaoshu Yu, Bo Wang, Pengpeng Zeng et al.· 0 citations
The Dynamic Buffer framework enables robots to form and transport temporary stacks of objects as single units that reduces total manipulator travel, can improve operational throughput, can reduce energy consumption, and provides flexible buffering where workspace is tight.
Arman Barghi, Hamed Hosseini, Seraj Ghasemi et al.· IEEE Transactions on Automat...· 3 citations
This paper introduces POEF (POlicy EFfective Jailbreak), an automated red-teaming framework that takes into account the robot-specific constraints during both the optimization and evaluation processes and proposes two defense strategies that mitigate the behavior jailbreak risks.
Xuancun Lu, Zhen Huang, Xin-Feng Li et al.· 17 citations· ⚡5
Reach audiences
Advertise in front of researchers, engineers, and readers.
A concept-centric framework for building agents that can learn continually and reason flexibly across multiple domains and offers several advantages, including data efficiency, compositional generalization, continual learning, and zero-shot transfer.
Jia-Yuan Mao, Joshua B. Tenenbaum, Jia-Jun Wu· Communications of the ACM· 14 citations· ⚡1
X-Reset is proposed, a framework that resolves exploration with human hand-object demonstrations of RL from scratch and scales with the number of training objects, generalizes to unseen objects, can learn from imperfect hand-pose estimates, and transfers behaviors zero-shot from sim-to-real.
Prithwish Dan, Chen-Yang Ma, Wei Zhan· 0 citations
Failure for Rising is proposed, a failure-driven real-to-sim-to-real closed-loop learning framework that converts real-world failures into targeted policy improvement and reconstructs each failure as an interactive, object-centric table-top environment that preserves the task-relevant spatial and physical conditions.
Zhuo-Yuan Yu, Jia-Cheng Wang, Tian-Le Liu et al.· 0 citations
CATok is proposed, a causal action tokenizer that reframes tokenization as a causally structured generative process, establishing a high-performance, scalable foundation for purely autoregressive VLA systems.
Chen-Yu Zhang, Yu-Hang Cao, Da-Ru Du et al.· 0 citations
P2P-T is introduced, from Pixel to Poses for Tool Manipulation, a data-efficient, object-centric framework that learns tool use directly from human demonstrations and achieves a 73% improvement over the previous state of the art in execution performance on complex, real-world tool manipulation tasks that currently rema...
Bang-Jun Wang, Long-Yan Wu, Yu-Kun Wei et al.· 0 citations
Spatial Grafting is proposed, a versatile, lightweight spatial module that binds frozen reconstruction features to metric, robot-relative geometry and injects them into the flow-matching action expert through cross-attention without modifying the host's perceptual pathway, so the host retains the full benefit of its pr...
Ding-Sheng Liu, Yang-Zheng Wu, Mahboubeh Asadi et al.· 0 citations
Comparisons within ReCAT show that the observation encoder and every-block memory conditioning are needed for this performance, and that update rules developed for efficient sequence modeling behave differently as robot memory: additive updates have the highest observed success on counting and timing, and delta-rule up...
Pankhuri Vanjani, M. Hatab, Can Mizrakli et al.· 0 citations
EMPIRIC, an agent that learns a residual world model: a physics engine extended with code for the missing mechanisms, that learns interpretable, reusable models, and solves more tasks with fewer environment interactions than all three baselines.
Yi-Chao Liang, Amber Li, Dat Nguyen et al.· 1 citation
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.
