arXiv：cs.AI· Wei Wang, Wenqiao Zhang, Yutong Lin, Yuqian Yuan, Tianwei Lin, Jinhao Mao, Zhenxuan Fan, Mingjian Gao, Yang Dai, Wentong Li, Zheqi Lv, Zheng Dong, Yingjie Niu, Jiaqi Zhu, Jun Xiao, Chao Li, Yueting Zhuang·· 6 小时前
EmbodiedSkills：用世界动作模型增强级联技能的 VLA 智能体框架
EmbodiedSkills: World Action Model-augmented Cascaded Skills for Vision-Language-Action Agent
AI 导读
EmbodiedSkills 是一个基于技能的智能体框架，通过世界动作模型为冻结的 VLA 策略增强后果感知的动作选择。在 MetaWorld MT50 和 CLIPort 上，成功率分别从 76.28% 提升至 78.64%、33.94% 提升至 36.83%；搭配冻结的 FLOWER 策略在 CALVIN 上五任务链成功率从 77.10% 升至 77.90%。
正文
Authors:Wei Wang, Wenqiao Zhang, Yutong Lin, Yuqian Yuan, Tianwei Lin, Jinhao Mao, Zhenxuan Fan, Mingjian Gao, Yang Dai, Wentong Li, Zheqi Lv, Zheng Dong, Yingjie Niu, Jiaqi Zhu, Jun Xiao, Chao Li, Yueting Zhuang
Abstract:Vision-language-action (VLA) policies generate executable robot actions, but plausible proposals can have different consequences for the task. We present EmbodiedSkills, a skill-based agent framework that augments frozen VLA policies with world action models for consequence-aware action selection. The framework defines reusable skills for scene observation, context encoding, candidate generation, consequence prediction, outcome evaluation, candidate comparison, and action execution. Five stages organize their input-output dependencies in a shared feedback loop, accommodating both continuous action chunks and pick-and-place primitives. At each decision point, independently trained world action models predict the latent consequences of a common set of policy proposals. Selection combines their outcome scores and prediction disagreement with a lightweight comparator that estimates the benefit of an alternative from agreement in its predicted effects. The predictive and comparison modules learn from matched-state candidate branches, while the action policy and pretrained observation encoders remain frozen. On MetaWorld MT50 and CLIPort, EmbodiedSkills improves success rates from 76.28% to 78.64% and from 33.94% to 36.83%, respectively. An additional CALVIN evaluation with the frozen FLOWER policy increases five-task-chain success from 77.10% to 77.90%. These results demonstrate the applicability of consequence-aware selection across different policy backends and action representations.
| Subjects: | Robotics (cs.RO); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2609.01281 [cs.RO] |
| (or arXiv:2609.01281v2 [cs.RO] for this version) | |
| https://doi.org/10.48550/arXiv.2609.01281 arXiv-issued DOI via DataCite |
Submission history
From: Wei Wang [view email]
[v1]
Tue, 1 Sep 2026 14:14:47 UTC (2,139 KB)
[v2]
Fri, 9 Oct 2026 06:31:30 UTC (2,339 KB)
来源：arXiv：cs.AI · arxiv.org