Robots in dynamic, human-centric environments must follow language instructions while maintaining real-time reactive control. Vision-language-action (VLA) models offer a promising framework, but they assume temporally aligned reasoning and control, despite semantic inference being inherently delayed relative to real-time action. We introduce Think-in-Control (TIC)-VLA, a latency-aware framework that explicitly models delayed semantic reasoning during action generation. TIC-VLA defines a delayed semantic-control interface that conditions action generation on delayed vision-language semantic states and explicit latency metadata, in addition to current observations, enabling policies to compensate for asynchronous reasoning. We further propose a latency-consistent training pipeline that injects reasoning inference delays during imitation learning and online reinforcement learning, aligning training with asynchronous deployment. To support realistic evaluation, we present DynaNav, a physics-accurate, photo-realistic simulation suite for language-guided navigation in dynamic environments. Extensive experiments in simulation and on a real robot show that TIC-VLA consistently outperforms prior VLA models while maintaining robust real-time control under multi-second reasoning latency. Project website: https://ucla-mobility.github.io/TIC-VLA/

翻译：在动态、以人为中心的环境中工作的机器人，必须遵循语言指令，同时保持实时的反应性控制。视觉-语言-动作（VLA）模型提供了一个有前景的框架，但它们假设推理与控制在时间上是对齐的，尽管语义推理本质上相对于实时动作存在延迟。我们提出了思控（Think-in-Control, TIC）-VLA，这是一个延迟感知框架，它在动作生成过程中显式地对延迟的语义推理进行建模。TIC-VLA定义了一个延迟的语义-控制接口，该接口除了基于当前观测外，还基于延迟的视觉-语言语义状态和显式的延迟元数据来生成动作，从而使策略能够补偿异步推理。我们进一步提出了一种延迟一致性训练流程，该流程在模仿学习和在线强化学习中注入推理延迟，使训练与异步部署保持一致。为了支持真实的评估，我们提出了DynaNav，这是一个物理精确、照片级真实的仿真套件，用于动态环境中的语言引导导航。在仿真和真实机器人上进行的大量实验表明，TIC-VLA始终优于先前的VLA模型，同时在多秒级推理延迟下保持稳健的实时控制。项目网站：https://ucla-mobility.github.io/TIC-VLA/