{"ID":6267250,"CreatedAt":"2026-07-10T01:11:38.759438437Z","UpdatedAt":"2026-07-13T01:02:08.706470581Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2607.08575","arxiv_id":"2607.08575","title":"FabriVLA: A Lightweight Vision-Language-Action Model for Precise Multi-Task Manipulation","abstract":"We present FabriVLA, a lightweight Vision-Language-Action model for Precise Multi-Task Manipulation. FabriVLA combines an InternVL3.5 vision-language backbone with a flow-matching action head featuring gated self-attention across action tokens and shallow VLM layer fusion for enriched spatial context. The model is trained via single stage joint optimization from a pretrained VLM and randomly initialized action head. On the Meta-World MT50 benchmark spanning 50 diverse manipulation tasks, FabriVLA achieves a tier-average success rate of 90.0%, demonstrating that a compact VLA built on a 1B scale VLM can achieve strong performance without relying on multi billion parameter VLA backbones.","short_abstract":"We present FabriVLA, a lightweight Vision-Language-Action model for Precise Multi-Task Manipulation. FabriVLA combines an InternVL3.5 vision-language backbone with a flow-matching action head featuring gated self-attention across action tokens and shallow VLM layer fusion for enriched spatial context. The model is trai...","url_abs":"https://arxiv.org/abs/2607.08575","url_pdf":"https://arxiv.org/pdf/2607.08575v1","authors":"[\"Shiyuan Yang\",\"Borong Zhang\",\"Jizheng Zhang\",\"Zhijia Tao\",\"Junfei Guo\",\"Donglai Ran\",\"Xu Bian\",\"Qingbiao Li\"]","published":"2026-07-09T15:06:43Z","proceeding":"cs.RO","tasks":"[\"cs.RO\"]","methods":"[]","has_code":false}