ReViP: Reducing False Completion in Vision-Language-Action Models with Vision-Proprioception Rebalance

作者：Yinghao Li, Jian-Jian Jiang, Lang Zhou, Tianyu Zhang, Jiadong Yin, Mu Lin, Yi-Lin Wei, Wei-Shi Zheng · 单位：Sun Yat-sen University, China, Shenzhen Loop Area Institute, China, Beijing Institute of Technology, China, Peng Cheng Laboratory, Shenzhen, China, Key Laboratory of Machine Intelligence and Advanced Computing, Ministry of Education, China, Guangdong Province Key Laboratory of Information Security Technology, China · 会议/期刊：arXiv · 日期：2026-01-23 · 来源：Low-Level Learning-Based Action Modelling / Input Modelling / 2D Vision Language Action Models with Auxiliary Tasks - Text Goal Extraction