JAM统一机器人控制研究路线
AI 总结
Yonghao Yu、Lang Huang、Runyi Li和Toshihiko Yamasaki在一篇论文中提出Just Action Model（JAM），通过预训练视觉骨干、动作专家、动作损失和前瞻损失构建统一框架。论文称，JAM策略在LIBERO和LIBERO-Plus上优于参数匹配的FastWAM与ImageWAM，在RoboTwin上取得相近或略高的成功率。使用缓存的前瞻目标后，JAM每次样本更新的估算在线训练FLOPs较FastWAM减少73%。
深度见解
论文提出将骨干初始化、未来观察目标和推理机制解耦，为比较视觉-语言-动作模型与世界-动作模型提供了一套具体框架。
Computer Science > Computer Vision and Pattern Recognition
arXiv:2610.37219 (cs)
[Submitted on 27 Oct 2026]
Title:Vision-Language-Action or World-Action? Just Action Models
Authors:Yonghao Yu, Lang Huang, Runyi Li, Toshihiko Yamasaki
View a PDF of the paper titled Vision-Language-Action or World-Action? Just Action Models, by Yonghao Yu and 3 other authors
#Just Action Model#FastWAM#ImageWAM#LIBERO#LIBERO-Plus#RoboTwin#Yonghao Yu#Lang Huang#Runyi Li#Toshihiko Yamasaki