Experiments show that VLDA improves long-horizon consistency, robustness, and generalization, validating actionable Delta as an alternative to direct action prediction and full future-state modeling.
Ling-Sen Zhang, Wei Li, Jie He et al.· 0 citations
ProAct is introduced, a world-calibrated proposal-to-action framework that makes the generative source itself predictable and improves performance across simulation and real-world tasks while reducing denoising steps by 50%, inference latency by up to 25.8%, and increasing throughput by up to 34.8%.
The introduction of ATI-VLA, an Action-Centric Predictive Vision-Language-Action framework via Actionable Alignment Then Adaptive Injection, which aligns predictive observation and action representations by mapping both modalities into a shared discrete latent space via a unified codebook, making predictive observation...
Vision-Language-Action (VLA) models have shown strong promise for general-purpose robotic manipulation, yet adapting them to new tasks and domains remains inefficient: existing methods often rely on parameter tuning, incurring substantial costs and risking catastrophic forgetting of previously learned tasks. To address...
Zai-Jing Li, Rui Shao, Bing Hu et al.· arXiv.org· 0 citations
Recent graphical user interface (GUI) grounders have significantly advanced single-shot accuracy on standard benchmarks, yet their performance degrades sharply on small targets, densely packed controls and out-of-distribution interfaces. We attribute this gap to a paradigmatic limitation shared by existing approaches:...
Renshan Zhang, Hao-Yang Meng, Yi-Xiao He et al.· 0 citations
A literature review around the interfaces connecting contextual evidence to execution, distinguishing four families: context-conditioned policies, geometric demonstration transfer, world-model-based control, and skill- and agent-based execution.
UniMPA, a Unified Memory-Prediction-Action model that addresses transition ambiguity by modeling the intended future state evolution through a shared action-grounded transition interface, is proposed.
World Action Models (WAMs) extend Vision-Language-Action (VLA) models by incorporating future visual dynamics into action generation. However, existing WAMs often utilize imagined futures with limited adaptation to evolving execution progress, potentially introducing distracting or unreliable predictive cues. This limi...
Yi-Jie Zhu, Zi-Tong Yu, Wei Li et al.· 1 citation
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.