发表机构
California Institute of Technology; New York University（加州理工学院; 纽约大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对文本条件运动生成器缺乏场景依赖安全感知的问题，提出无需训练的上下文安全过滤（CSF），可降低危险事件率达90%，在Unitree G1上成功阻止多种场景下的不安全运动。
AI 中文摘要
文本条件运动生成器可生成可追踪的全身运动，但不具备场景依赖的安全感知能力：同一动作可能针对物体或人。现有安全措施要么检查提示词、需要带标注的运动数据，要么强制执行几何约束，因此无法直接考虑场景上下文如何改变运动的含义。我们提出上下文安全过滤（CSF），这是一种无需训练的过滤器，它将自然语言安全规则基于生成器产生的安全与不安全参考轨迹。对于每条激活的规则，安全与不安全参考轨迹定义了一个仿射安全值，由安全参考跟踪CBF-QP强制执行。在四个具有不同架构的预训练生成器上，CSF在所有明确及场景触发的不安全案例中激活了预期规则，将危险事件率降低了多达90%，同时保留了88%-100%的良性运动。我们在真实世界的Unitree G1上展示了完整系统，其成功在多种场景中阻止了不安全运动，包括与人类和物体的交互。
英文摘要
Text-conditioned motion generators produce trackable whole-body motion, but they have no notion of scene-dependent safety: the same action may target an object or a person. Existing safeguards either inspect the prompt, require labeled motion data, or enforce geometric constraints; therefore, they do not directly account for how scene context changes a motion's meaning. We introduce contextual safety filtering (CSF), a training-free filter that grounds natural-language safety rules in safe and unsafe reference trajectories produced by the generator. For each active rule, safe and unsafe reference trajectories define an affine safety value that a safe reference tracking CBF-QP enforces. Across four pretrained generators with different architectures, CSF activates the intended rules in all explicit and scene-triggered unsafe cases and reduces the danger-event rate by up to 90%, while preserving 88-100% of benign motions. We demonstrate the complete system on a real-world Unitree G1, where it successfully prevents unsafe motions in a variety of scenarios, including interactions with humans and objects.
Comments8 pages, 6 figures, website at https://lzyang2000.github.io/csf/
