This work proposes DreamAvoid, a critical-phase test-time dreaming framework that enables VLA models to anticipate and avoid failures, and introduces an autonomous boundary learning paradigm to refine the system's understanding of the subtle boundary between success and failure.
Shape-Aware Reinforcement Learned Model Predictive Control is proposed, a method for safe, efficient, and adaptive navigation in crowds with heterogeneous shapes without geometry simplification that preserves the safety structure and generalizability of MPC while integrating the adaptability and intelligence of RL.
StreamPI is proposed, a streaming multimodal temporal modeling framework that equips single-frame VLA with temporal reasoning capability without introducing any additional parameters and seamlessly inherits pretrained single-frame weights and supports flexible single-frame and multi-frame inference.
Zhe Liu, Jinghua Hou, Yuxiang Lu et al.· 1 citation· ⚡1
SpectraReward is proposed, a training-free reward function that turns pretrained MLLMs into off-the-shelf reward models for image-generation reinforcement learning, and Self-SpectraReward is introduced, a special case for unified multimodal models where the policy's own understanding branch serves as the reward model f...
Runhu Huang, Qihui Zhang, Zhe Liu et al.· arXiv.org· 0 citations
ACE-Brain-0.5 is presented, a unified embodied foundation model that organizes robot intelligence into five coupled functions: spatial perception, decision making, embodied interaction, self-monitoring, and self-improvement, and SSR+, which extends Scaffold-Specialize-Reconcile with a Reactivate stage after task-vector...
Zi-Yang Gong, Hao-Ming Gu, Ze-Hang Luo et al.· arXiv.org· 3 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
