This work proposes CF-VLA, a coarse-to-fine two-stage formulation that re-structures action generation into a coarse initialization step that constructs an action-aware starting point, followed by a single-step local refinement that corrects residual errors.
IMLE-VLA is introduced, which replaces the iterative action head with a single-step conditional generator trained via conditional Implicit Maximum Likelihood Estimation (cIMLE), and promotes multimodal action coverage, avoiding the mode collapse of naive regression heads while eliminating multi-step sampling entirely.
Kian Hosseinkhani, Qin-He Peng, George Shramko et al.· arXiv.org· 2 citations
Conventional flow-based vision-language-action (VLA) models support expressive continuous action generation but rely on multi-step refinement to produce each action chunk, increasing latency in online robot control. To address this issue, we introduce DriftingVLA, a native one-step VLA that generates a complete action...
TEMPO is proposed, a semantic-action decoupled, two-timescale RL post-training framework for VLA models that consistently outperforms both pretrained state-of-the-art VLA models and the RL post-training baseline, while reaching and maintaining higher evaluation rewards on two real-world tasks.
Vision-Language-Action (VLA) models have shown strong generalization in robotic manipulation by combining semantic knowledge from pretrained vision-language models with expressive action-generation policies. Diffusion-based action generators are particularly effective for modeling temporally coherent action chunks, but...
Yi-Heng Ji, Xing-Ru Zhou, Luis Sentis et al.· 0 citations
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.
Results indicate that a fixed VLA policy can benefit from conditioning on spatiotemporally aligned demonstrations at test time, and present ICI-VLA, a training and retrieval framework that equips a text-action VLM with few-shot test-time adaptation through in-context demonstrations.
Song-Hua Yang, Zi-Yu Liu, Xue-Tao Li 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.