{"ID":5937635,"CreatedAt":"2026-07-07T03:14:33.014478982Z","UpdatedAt":"2026-07-08T08:23:32.674931315Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2607.04171","arxiv_id":"2607.04171","title":"XS-VLA: Coupling Coarse-grained Spatial Distillation with Latent Flow Matching for Lightweight Robotic Control","abstract":"Large Vision-Language Models (LVLMs) have shown strong multimodal understanding and spatial grounding, but their computational cost limits real-time robotic control. In contrast, lightweight models are suitable for edge deployment but often suffer from \"spatial blindness\", namely weak native spatial prediction ability. Training Vision-Language-Action (VLA) models on mixed human demonstrations can also degrade policy performance due to highly diverse behaviors. To address these limitations, we propose XS-VLA, a two-stage framework for efficient and spatially grounded robotic manipulation. First, we distill spatial semantic knowledge from Qwen3-VL-4B into the SmolVLM2-0.25B backbone by fine-tuning on curated coarse-grained spatial descriptions, turning the lightweight model into a spatially grounded engine. Second, we use this enhanced backbone to condition a Latent Flow Matching policy. Unlike deterministic controllers, our policy combines a Conditional Variational Autoencoder (CVAE) with Flow Matching dynamics to model complex multimodal action distributions. On the LIBERO benchmark, XS-VLA achieves state-of-the-art performance among models with fewer than 0.5B parameters. It improves average success rates by up to 7.2 percent, including a 23 percent gain on LIBERO-Long, over the SmolVLA 0.25B baseline, and outperforms the larger 2.2B vanilla SmolVLA. Ablations show that spatial tuning and generative latent flow control substantially improve lightweight VLA performance, delivering a 3.2 times speedup in mission execution over the previous lightweight flow matching policy.","short_abstract":"Large Vision-Language Models (LVLMs) have shown strong multimodal understanding and spatial grounding, but their computational cost limits real-time robotic control. In contrast, lightweight models are suitable for edge deployment but often suffer from \"spatial blindness\", namely weak native spatial prediction ability....","url_abs":"https://arxiv.org/abs/2607.04171","url_pdf":"https://arxiv.org/pdf/2607.04171v1","authors":"[\"Lei Iok Tong\",\"Qingchen Xie\",\"Wei Huang\",\"Ying Jie Yap\",\"Yujie Zhang\",\"Qianzhi Li\",\"Xiaolong Liu\",\"Zhidong Deng\"]","published":"2026-07-05T08:34:39Z","proceeding":"cs.RO","tasks":"[\"cs.RO\",\"cs.LG\"]","methods":"[\"Language Model\",\"Variational Autoencoder\"]","has_code":false}