IntentVLA is introduced, a history-conditioned VLA framework that encodes recent visual observations into a compact short-horizon intent representation and uses it to condition chunk generation and improves rollout stability and outperforms strong VLA baselines.
Shijie Lian, Bin Yu, Xiaopeng Lin et al.· arXiv.org· 6 citations
The core design principle of DELE-w0.5 is to model how the physical world changes under robot actions, rather than how its visual appearance evolves frame by frame, which enables cheaper training and low-latency inference.
Fenghao Lei, Zhixiong Huang, Long Yang et al.· 2 citations
Experimental results show that action filtering provides negligible safety improvement, while observation filtering reduces near mid-air collisions by 90% and remains robust to the barrier function's tradeoff between separation distance and closing rate, and suggest that preserving the policy's decision authority outpe...
Experiments show that J-LAW's factor-graph representation can improve long-horizon latent consistency and recover more reliable predictive states under partial observations, forming a foundation for future integrated localization and planning systems.
Guanqun Cao, Liang Chen· 0 citations
Reach audiences
Advertise in front of researchers, engineers, and readers.
This work derives a closed-form expression for this adversarial perturbation, bypassing the iterative inner optimization of adversarial training entirely and enabling linear-time evaluation in the state dimension, and shows that this expression approximates the exact minimizer of the value function over the modeled unc...
Alex Zongo, Filippos Fotiadis, U. Topcu et al.· arXiv.org· 1 citation
A fixed encoder is studied in a fixed encoder and trajectory reachability metrics (TRM), a small temporal pairwise cost trained from logged trajectories and used to rank predicted endpoints of a candidate action sequence against a goal is introduced.
Lian Li, Shengzhi Wang, Li-Bin Qiu et al.· 2 citations
This work introduces Constrained GRPO, a Lagrangian-based extension of GRPO for constrained policy optimization, and addresses the coupling induced by reward scalarization by scalarizing standardized advantages rather than rewards.
Roger Girgis, Rodrigue de Schaetzen, Luke Rowe et al.· arXiv.org· 2 citations· ⚡1
This work removes a model's camera input and replaces it with memories from prior drives at the same location, suggesting that a high NAVSIM score does not require a planner to react to the current traffic scene and should be treated with caution.
Christian Löwens, Thorben Funke, Alexandru Condurache· 0 citations
Motus2 is presented, a self-evolving general world model for dexterous manipulation that combines egocentric data scaling and closed-loop general world model scaling to provide a general path toward self-evolving dexterous manipulation.
Hong-Zhe Bi, Zikun Zhou, Yihao Tang et al.· 7 citations
A paradigm for tactile-enabled embodied manipulation, which integrates tactile sensing hardware, large-scale multimodal data, tactile representation learning, and standardized evaluation is presented, aiming at supporting future work on tactile-enabled embodied manipulation.
A novel metric derived from the flow matching trajectory curvature is introduced to quantify action generation confidence during inference and enables the dynamic reduction of inference steps and the adaptive adjustment of MLP pruning ratios through an efficiently computed importance evaluation, requiring no access to...
A proof-of-concept Brain-Language-Action (BLA) model for drone control using motor-imagery EEG from the BCI Competition IV 2a dataset is developed, providing an initial demonstration that language conditioning can expand the effective control range of EEG-based robotic interfaces without requiring a corresponding incre...
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.
