World-Action Models inherit world knowledge from video-generative priors, and channel it into executable control signals through embodied experience. Existing systems, however, are monolithic: the generative backbone, visual representation, architecture, information flow, inference procedure, and training data are tigh...
Yu-Ran Wang, Siqiao Huang, Ming-Lei Li et al.· 38 citations· ⚡4
Precise grounding matters. It specifies which object is the target and where that object is, even in clutter and for tiny objects, and it has to be fast enough for closed-loop control. Yet vision-language-action (VLA) and world-action models (WAMs) take perception from general-purpose vision-language and video-generati...
Generalist robot policies benefit from large-scale visual pretraining but often struggle with contact-rich dexterous manipulation, where force, compliance, and incipient slip are hard to observe from images. We present Spark-0, a spatially aligned vision-tactile-language-action (VTLA) model that uses touch in two compl...
Tian-Xing Chen, Wen-Wei Lin, Chuqiao Lyu et al.· 0 citations
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
Embodied AI systems are often organized into System 1 and System 2. System 1 is typically a pretrained policy that generates actions at high frequency, whereas System 2 is often instantiated as a vision-enabled language model for high-level planning. We ask whether a large language model (LLM) can act as the policy for...
Learning predictive models of contact-rich dexterous manipulation requires dense tactile interaction, but such data are costly to scale on real robots and remain tied to embodiment-specific sensors. We introduce DexTouch-WM, an action-conditioned world model that learns from scalable human touch to jointly predict futu...
Yan-Jiao Qin, Yue Chen, Wen-Wei Lin et al.· 3 citations
Precise grounding matters. It specifies which object is the target and where that object is, even in clutter and for tiny objects, and it has to be fast enough for closed-loop control. Yet vision-language-action (VLA) and world-action models (WAMs) take perception from general-purpose vision-language and video-generati...
Qi-Ze Yu, Lian-Rui Fan, Bo-Yu Chen et al.· 0 citations
AffordanceWAM is introduced, an affordance-aware generative World Action Model that represents object-centric spatiotemporal affordance through Scalar Affordance and Affordance Heatmap, within the generated future World, and supports affordance as an effective interface for both vision-language-action learning and huma...
HeteroGenManip is proposed, a task-conditioned, two-stage framework designed to decouple initial grasp from complex interaction execution, and achieves robust intra-category shape and pose generalization.
From YAML-first specifications that decouple contents, placement, behavior, and agent exposure, MagicSim constructs diverse executable worlds spanning task families, interaction regimes, physics, layouts, sensors, avatars, and robot embodiments in one reset-and-step loop.
Xpolicylab is released as shared infrastructure for reproducible policy comparison and standardized deployment across simulation and physical platforms for reproducible policy comparison and standardized deployment across simulation and physical platforms.
