arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2602.14534cs.CV
收藏

MoRL: 用于统一运动理解与生成的强化推理

MoRL: Reinforced Reasoning for Unified Motion Understanding and Generation

The University of Sydney(悉尼大学);Peking University(北京大学);Nanyang Technological University(南洋理工大学)

Hongpeng Wang, Zeyu Zhang, Wenhao Li, Hao Tang

AI总结:

MoRL通过结合监督微调和强化学习,提升运动理解与生成的推理能力和现实感,并引入链式运动方法和大规模CoT数据集以增强推理效果。

摘要:

人类运动理解与生成对于视觉和机器人领域至关重要,但仍然在推理能力和测试时规划方面存在局限。我们提出了MoRL,一种统一的多模态运动模型,通过监督微调和具有可验证奖励的强化学习进行训练。我们的任务特定奖励设计结合了语义对齐和推理一致性以用于理解,以及物理合理性和文本-运动一致性以用于生成,从而提高了逻辑推理和感知现实感。为进一步增强推理,我们引入了链式运动(CoM),一种测试时推理方法,能够实现逐步规划和反思。我们还构建了两个大规模的CoT数据集,MoUnd-CoT-140K和MoGen-CoT-140K,以将运动序列与推理轨迹和动作描述对齐。在HumanML3D和KIT-ML上的实验表明,MoRL在现有最先进基线之上取得了显著提升。代码:https://github.com/AIGeeksGroup/MoRL。网站:https://aigeeksgroup.github.io/MoRL。

英文摘要:

Human motion understanding and generation are crucial for vision and robotics but remain limited in reasoning capability and test-time planning. We propose MoRL, a unified multimodal motion model trained with supervised fine-tuning and reinforcement learning with verifiable rewards. Our task-specific reward design combines semantic alignment and reasoning coherence for understanding with physical plausibility and text-motion consistency for generation, improving both logical reasoning and perceptual realism. To further enhance inference, we introduce Chain-of-Motion (CoM), a test-time reasoning method that enables step-by-step planning and reflection. We also construct two large-scale CoT datasets, MoUnd-CoT-140K and MoGen-CoT-140K, to align motion sequences with reasoning traces and action descriptions. Experiments on HumanML3D and KIT-ML show that MoRL achieves significant gains over state-of-the-art baselines. Code: https://github.com/AIGeeksGroup/MoRL. Website: https://aigeeksgroup.github.io/MoRL.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑