World action models (WAMs) have emerged as an important class of generalist robot policies, aiming to transfer video prediction priors to action learning. However, we find that existing WAMs frequently produce incorrect motion predictions in open-ended environment, leading to erroneous actions. We attribute this limita...
Reliable robotic manipulation requires both semantic understanding and precise control in three-dimensional space. However, vision-language-action (VLA) foundations largely inherit visual representations from 2D vision-language pre-training, while spatial supervision is often introduced only during task-specific adapta...
Fu-Hao Li, Wen-Xuan Song, Qi-Zhou Huang et al.· 0 citations
JADE is introduced, a Joint Action DEnoiser that enables the VLM and action expert to jointly refine actions and enables JADE to generate actions with fewer sampling steps, helping offset the additional cost of VLM participation.
Xi-Cheng Gong, Wen-Xuan Song, Cheng-Ze Liu et al.· 0 citations
Recent Vision-Language-Action (VLA) models increasingly use natural-language instructions to condition autonomous driving behavior. Yet existing closed-loop benchmarks primarily measure route completion, safety, and rule compliance, leaving a fundamental question unresolved: does language actually determine which feasi...
Zi-Han You, Ze-Hua Fan, Wenqi Lyu et al.· 0 citations
Action-chunking policies predict multi-step control sequences, but a fundamental question remains: how much of a predicted action chunk should be committed before replanning? Existing systems typically execute a fixed-length prefix, implicitly assuming that the same execution horizon remains trustworthy across states....
Yu-Yan Li, Yu-Jian Wang, Yu-Song Huang et al.· 0 citations
Scaling robotic foundation models requires diverse training data and reliable evaluation environments. Simulation offers a scalable solution, yet existing generation pipelines remain constrained by predefined assets and skills, a disconnect between scene generation and task generation, and limited support for complex e...
Yi-Kai Qin, Yi-Fei Deng, Ming-Jian Liang et al.· 0 citations
Most vision-language-action (VLA) models rely on current observations and lose task-relevant evidence once it leaves view, limiting performance on long-horizon, memory-dependent tasks. Existing efforts incorporate compressed historical features or sparse visual keyframes. However, isolated snapshots can leave the polic...
Yu-Heng Na, Zhi-De Zhong, Jun-Jie He et al.· 1 citation
Robotic data generation is a promising paradigm for scaling robot learning without collecting large-scale real-world data. However, generating geometrically diverse yet physically valid data for contact-rich tasks remains challenging, especially when success depends on precise geometric interfaces. Standard shape augme...
Tian-Yi Xiang, Xupeng Xie, Jia-Hang Cao et al.· 0 citations
UniWAM is introduced, a unified architecture that integrates a physical reasoner, a world generator, and an action predictor to jointly learn semantic understanding of the physical world, visual generation, and action prediction and uncovers a log-linear scaling law of unified human-robot co-training.
Wen-Xuan Song, Jia-Yi Chen, Jing-Bo Wang et al.· 6 citations· ⚡2
World-Action Models (WAMs) couple visual dynamics with action prediction, bringing the rich priors of pretrained video models to robotic manipulation. However, their multi-view interfaces typically tile images or concatenate tokens, leaving the geometric relationships among synchronized cameras implicit. This makes it...
SLIP-VLA, a policy learning framework that equips VLA models with a Single-Step Latent Imagination for future-aware action prediction, achieves state-of-the-art performance across diverse simulation benchmarks and real-world manipulation tasks, while its single-step latent imagination takes only 12 ms.
This work designs a hybrid architecture that combines sparse anchors with an offset refinement module for efficient multi-modal trajectory prediction and adopts a two-stage training paradigm consisting of pretraining and reward-guided fine-tuning, demonstrating superior performance under real-time constraints.
Cheng-Lin Chen, Lu-Jia Wang, Xin-Hu Zheng et al.· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.