机器人VLA模型从实验室走向实用：三大改进
机器人VLA（视觉-语言-动作）模型在实验室表现不错，但一到真实世界就掉链子。这篇论文从三个方向动手改进：一是用6万小时数据（含5万小时20种机器人轨迹和1万小时人类第一人称视频）训练，让模型能跨任务、跨机器人形态泛化；二是扩展动作空间，让机器人能控制头、腰、移动底座和灵巧手，完成更复杂的任务；三是加入预测动力学建模，通过视频和深度估计预测未来状态，提升时间推理能力。在GM-100基准测试中，这些改进带来了显著提升。虽然你明天用不上，但它展示了VLA模型从实验室走向实用的关键路径。
📄 原文摘要(英文)
Despite recent progress of VLA foundation models, the disparity between laboratory conditions and real-world applications continues to impede their practical implementation. To bridge this gap, we present LingBot-VLA 2.0, which advances LingBot-VLA through improvements in three functional domains. (1) Generalization across tasks and embodiments. Compared to the previous version, we revamp the data processing pipeline and curate around 60,000 hours of data for pretraining, including 50,000 hours of robot trajectories spanning 20 robot configurations and 10,000 hours of egocentric human videos. (2) Expanded action space in addition to dual-arm hardware platforms. In particular, our system accommodates degrees of freedom for the heads, waists, mobile bases, and dexterous hands, thereby empowering the robots to tackle more complex tasks in practical scenarios. (3) Predictive dynamics modeling for improved temporal reasoning. Specifically, we formulate future prediction as a proxy task, facilitated by a video representation model for semantic priors and a depth estimation model for geometric cues. Evaluations on the GM-100 benchmark, conducted in a generalist setting, validate the beneficial impact of these proposed modifications. Furthermore, benefiting from the expanded pretraining data that covers whole-body degrees of freedom, LingBot-VLA-2.0 demonstrates strong cross-embodiment long-horizon mobile manipulation capability across the two robotic platforms.