ACoT-VLA：面向视觉-语言-动作模型的动作思维链
ACoT-VLA: Action Chain-of-Thought for Vision-Language-Action Models
将思维链从语言/视觉空间迁移到动作空间，用粗动作意图序列作为中间推理步骤来条件化下游动作头，为视觉-语言-动作策略提供了一种可复用的推理增强范式。
原文依据 we posit that the most effective form of reasoning is one that deliberates directly in the action space
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论视觉-语言-动作模型已成为面向多样化操作任务的重要通用机器人策略，其通常依赖视觉-语言模型嵌入将多模态输入直接转换为动作。近期进展引入了显式的中间推理——例如子任务预测（语言）或目标图像合成（视觉）——来引导动作生成。然而，这些中间推理往往是间接的，并且在其传达精确动作执行所需的完整、细粒度信息的能力上存在固有局限。相反，我们认为最有效的推理形式是直接在动作空间中进行审慎思考。我们引入动作思维链（ACoT），一种将推理过程本身表述为引导最终策略的结构化粗动作意图序列的范式。在本文中，我们提出ACoT-VLA，一种实现ACoT范式的新颖架构。具体而言，我们引入两个互补组件：显式动作推理器（EAR）和隐式动作推理器（IAR）。前者提出粗参考轨迹作为显式动作级推理步骤，后者从多模态输入的内部表示中提取潜在动作先验，共同构成一个ACoT，以条件化下游动作头，从而实现有依据的策略学习。在真实世界和仿真环境中的大量实验证明了我们方法的优越性。代码见：https://github.com/AgibotTech/ACoT-VLA。
英文原文摘要 1562 字符
Vision-Language-Action models have emerged as essential generalist robot policies for diverse manipulation tasks, conventionally relying on directly translating multimodal inputs into actions via Vision-Language Model embeddings. Recent advancements have introduced explicit intermediary reasoning--such as sub-task prediction (language) or goal image synthesis (vision)--to guide action generation. However, these intermediate reasoning are often indirect and inherently limited in their capacity to convey the full, granular information required for precise action execution. Instead, we posit that the most effective form of reasoning is one that deliberates directly in the action space. We introduce Action Chain-of-Thought (ACoT), a paradigm where the reasoning process itself is formulated as a structured sequence of coarse action intents that guide the final policy. In this paper, we propose ACoT-VLA, a novel architecture that materializes the ACoT paradigm. Specifically, we introduce two complementary components: an Explicit Action Reasoner (EAR) and Implicit Action Reasoner (IAR). The former proposes coarse reference trajectories as explicit action-level reasoning steps, while the latter extracts latent action priors from internal representations of multimodal input, co-forming an ACoT that conditions the downstream action head to enable grounded policy learning. Extensive experiments in real-world and simulation environments demonstrate the superiority of our proposed method. Code is available at: https://github.com/AgibotTech/ACoT-VLA.