通过消除暴力动作生成，探索如何让模型‘遗忘’特定运动模式。
Human Motion Unlearning
- 用隐空间代码替换法，在不重新训练的情况下移除暴力动作。
- 在人类动作数据集上实现7%-15%暴力动作的有效抑制。
- 适合关注生成安全性的动作合成研究者与应用开发者。
我们提出人类动作遗忘（Human Motion Unlearning），以防止3D动作生成中的暴力行为，这是重要安全需求。当前主流文本到动作数据集（HumanML3D 和 Motion-X）中包含7%至15%的暴力序列，涵盖单个动作（如挥拳）和复合动作（如抬腿踢击）。通过聚焦暴力遗忘，我们验证了消除复杂多样的概念可作为运动“遗忘”能力的代理指标。为此，我们首次建立动作遗忘基准，自动筛选 HumanML3D 与 Motion-X 数据集，构建独立的遗忘集（暴力动作）与保留集（安全动作）。设计面向序列化遗忘的评估指标，衡量抑制效果及真实感与流畅性保持。将两种先进的免训练图像遗忘方法（UCE 与 RECE）适配至主流文本到动作架构（MoMask 与 BAMM），并提出新方法 Latent Code Replacement（LCR），在离散代码本中识别暴力代码并替换为安全替代项。实验表明，无需重新训练即可有效抑制暴力动作，且在隐空间操作实现了最佳质量平衡。本工作为多样化应用中的安全动作合成奠定基础。
原文摘要 · Abstract (English)
We introduce Human Motion Unlearning and motivate it through the concrete task of preventing violent 3D motion synthesis, an important safety requirement given that popular text-to-motion datasets (HumanML3D and Motion-X) contain from 7\% to 15\% violent sequences spanning both atomic gestures (e.g., a single punch) and highly compositional actions (e.g., loading and swinging a leg to kick). By focusing on violence unlearning, we demonstrate how removing a challenging, multifaceted concept can serve as a proxy for the broader capability of motion "forgetting." To enable systematic evaluation of Human Motion Unlearning, we establish the first motion unlearning benchmark by automatically filtering HumanML3D and Motion-X datasets to create distinct forget sets (violent motions) and retain sets (safe motions). We introduce evaluation metrics tailored to sequential unlearning, measuring both suppression efficacy and the preservation of realism and smooth transitions. We adapt two state-of-the-art, training-free image unlearning methods (UCE and RECE) to leading text-to-motion architectures (MoMask and BAMM), and propose Latent Code Replacement (LCR), a novel, training-free approach that identifies violent codes in a discrete codebook representation and substitutes them with safe alternatives. Our experiments show that unlearning violent motions is indeed feasible and that acting on latent codes strikes the best trade-off between violence suppression and preserving overall motion quality. This work establishes a foundation for advancing safe motion synthesis across diverse applications. Website: https://www.pinlab.org/hmu.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。