面向机器人操作的语言锚定解耦动作表征
Language-Grounded Decoupled Action Representation for Robotic Manipulation
把高层视觉-语言表征与低层动作控制解耦，用语言作为语义桥梁，并以翻译/旋转/夹爪三类动作基元作为细粒度中间层；对做 VLA 或视觉-语言条件策略的研究者，这套「语义引导软标签对比学习 + 课程式自适应权重」可直接迁移到自己的动作头对齐与跨任务泛化上。
原文依据 a semantic-guided soft-label contrastive learning objective to align similar action primitives across tasks
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论高层视觉-语言理解与低层动作控制之间的异构性，仍是机器人操作中的一个根本性挑战。尽管近期方法在任务特定的动作对齐上取得了进展，但它们往往难以为新颖任务或语义相关任务生成稳健而准确的动作。为此，我们提出语言锚定解耦动作表征（LaDA）框架，以自然语言作为语义桥梁连接感知与控制。LaDA 引入由三个可解释动作基元——平移、旋转与夹爪控制——构成的细粒度中间层，为低层动作提供显式的语义结构。它进一步采用语义引导的软标签对比学习目标，在任务之间对齐相似的动作基元，从而增强泛化能力与运动一致性。受课程学习启发的自适应加权策略动态平衡对比目标与模仿目标，以实现稳定而有效的训练。在仿真基准（LIBERO 与 MimicGen）以及真实世界演示上的大量实验验证，LaDA 取得了强劲性能，并能有效泛化到未见任务或相关任务。
英文原文摘要 1245 字符
The heterogeneity between high-level vision-language understanding and low-level action control remains a fundamental challenge in robotic manipulation. Although recent methods have advanced task-specific action alignment, they often struggle to generate robust and accurate actions for novel or semantically related tasks. To address this, we propose the Language-Grounded Decoupled Action Representation (LaDA) framework, which leverages natural language as a semantic bridge to connect perception and control. LaDA introduces a fine-grained intermediate layer of three interpretable action primitives--translation, rotation, and gripper control--providing explicit semantic structure for low-level actions. It further employs a semantic-guided soft-label contrastive learning objective to align similar action primitives across tasks, enhancing generalization and motion consistency. An adaptive weighting strategy, inspired by curriculum learning, dynamically balances contrastive and imitation objectives for stable and effective training. Extensive experiments on simulated benchmarks (LIBERO and MimicGen) and real-world demonstrations validate that LaDA achieves strong performance and generalizes effectively to unseen or related tasks.