FedVLA: Federated Vision-Language-Action Learning with Dual Gating Mixture-of-Experts for Robotic Manipulation

作者：Cui Miao, Tao Chang, Meihan Wu, Hongbin Xu, Chun Li, Ming Li, Xiaodong Wang · 单位：National University of Defense Technology, Bytedance Seed, University of Siedlce, Shenzhen MSU-BIT University, Shenzhen Technology University, Guangdong Laboratory of Artificial Intelligence and Digital Economy (SZ) · 会议/期刊：ICCV 2025 · 方向：专家混合 · 发布日期：2025-08-04