Zero-WAM introduces a causal video-action model that enables robots to perform unseen manipulation tasks by following in-context human video guidance. The authors create HumanGen, a dataset of 74.2K human-robot ICL pairs across 8.6K tasks, and propose an in-context future chunk prediction objective to prevent shortcut learning. In simulation, Zero-WAM attains a 47.0% success rate on seven unseen tasks, outperforming the best video-action baseline by 29.5 percentage points, and demonstrates real‑world generalization to complex, long‑horizon, and fine‑grained tasks.
By Jiaming Zhou, Qihang Zhang, Gangwei Xu, Cunxin Fan, Yujie Zhao, Ruilin Wang, Yiming Luo, Shuai Yang, Xing Zhu, Yujun Shen, Junwei Liang, Yinghao Xu
ActiveScale is a framework that enhances active perception for robots by integrating model, data, and hardware innovations. It augments vision‑language‑action models with historical video observations and explicit camera‑pose supervision, and introduces a scalable human‑robot mid‑training recipe using 1000 hours of egocentric and robotic data. The Active‑perception Mobile‑manipulation Platform (AMP) enables single‑operator teleoperation for scalable demonstration collection, leading to improved success rates on active‑perception tasks.
By Shuai Zhou, Kaisheng Pang, Wenxuan Song, Wenjie Zhang, Xinhu Zheng, Haoang Li
Zero-shot cross-task generalization, where a policy must execute manipulation tasks never seen during training, remains a central challenge in robot learning. In large language models, a novel task ca...
The paper introduces CamVLM, a framework that equips large vision‑language models with the ability to actively control camera viewpoints for improved surveillance video understanding. It presents two new datasets: CCTV‑Anomaly, a large‑scale surveillance video collection with detailed captions and event annotations, and CamTrack‑53K, an object‑centric viewpoint trajectory dataset for learning camera actions. Using reinforcement learning, CamVLM learns long‑horizon observation strategies, achieving state‑of‑the‑art performance in both passive and dynamic viewpoint settings.
By Xiao Zhang, Wang Zeng, Sheng Jin, Wentao Liu, Chen Qian, Shichao Kan
arXiv:2606. 00054v1 Announce Type: cross Abstract: Recent progress in generalizable embodied control has been driven by large-scale pretraining of Vision-Language-Action (VLA) models.
By Zhiyuan Feng, Qixiu Li, Huizhi Liang, Rushuai Yang, Yichao Shen, Zhiying Du, Zhaowei Zhang, Yu Deng, Li Zhao, Hao Zhao, Zongqing Lu, Oier Mees, Marc Pollefeys, Jiaolong Yang, Baining Guo
Generative video models can serve as a promising backbone for robot navigation by predicting future observations as video plans. Recent approaches often condition video planning on short-horizon guida...
arXiv:2609.22332v1 Announce Type: cross
Abstract: Generalizable robot manipulation requires predicting how a scene will evolve, identifying where interactions are feasible, and determining how to act...
By Jiadi You, Qize Yu, Yue Chen, Minghong Cai, Zhide Zhong, Yuran Wang, Bowen Ping, Jiaqi Liang, Zhenhao Shen, Haodong Yan, Yinchuan Li, Ruihai Wu, Xiaojuan Qi, Yingcong Chen
CueNav is a video model-based navigation framework that uses visual cues—a Bird's-Eye View map for global task context and a body-aware egocentric view for embodiment context—to guide a video planner. The framework couples this planner with an embodiment-specific Inverse-Dynamics Model that translates dense flow fields from the video plan into robot actions. Experiments show that CueNav nearly doubles maze navigation success compared to cue-less planning and achieves 70% success in narrow passages, while also supporting zero-shot semantic-conditioned navigation across different robot platforms.
By Hojin Lee, Sizhe Lester Li, Maximilian Hilger, Susie Lu, Achim J. Lilienthal, Vincent Sitzmann, Daniel A. Duecker
MINT is a foundation model that directly predicts world-space two-hand trajectories from egocentric RGB video, jointly estimating camera motion, hand states, and hand presence in a single spatiotemporal representation. It uses an open-source labeling pipeline, EGOPIPELINE, to generate large-scale pseudo-labels for pretraining, followed by fine-tuning on a small set of high-quality joint annotations. The model outperforms existing multi-stage approaches in accuracy and speed, and generalizes zero‑shot to unseen egocentric datasets.
By Zijie Zhu, Weiren Cai, Yizhou Wang, Zhenjie Yang, Yide Liu, Jiahao Chen, Guanqi He
The paper introduces Intention Distillation (INDI), a method that injects behavior-level intent into Vision‑Language‑Action (VLA) model decoders by leveraging a frozen teacher vision‑language model to interpret demonstrations. During training, the teacher processes the current observation, instruction, coarse action summary, and execution video, producing a multimodal intent representation that the VLA decoder uses alongside trajectory and execution features to predict actions. Experiments on SimplerEnv‑Bridge, RoboCasa Kitchen, and real‑world tasks show that INDI consistently improves success rates, especially on longer‑horizon tasks, demonstrating that explicit modeling of semantic intent benefits action decoders.
By Sangoh Lee, Sangwoo Mo, Wook-Shin Han
arXiv:2605. 30280v2 Announce Type: replace-cross Abstract: Embodied intelligence is often studied through specialized models for individual tasks such as manipulation or navigation, resulting in fragmented capabilities and limited generalization across tasks, environments, and robot embodiments.
By Qiuyue Wang, Mingsheng Li, Jian Guan, Jinhui Ye, Sicheng Xie, Yitao Liu, Junhao Chen, Zhixuan Liang, Jie Zhang, Xintong Hu, Xuhong Huang, Pei Lin, Junyang Lin, Dayiheng Liu, Shuai Bai, Jingren Zhou, Jiazhao Zhang, Haoqi Yuan, Gengze Zhou, Hang Yin, Ye Wang, Yiyang Huang, Zixing Lei, Wujian Peng, Delin Chen, Yingming Zheng, Jingyang Fan, Xianwei Zhuang, Xin Zhou, Haoyang Li, Anzhe Chen, Tong Zhang, Xuejing Liu, Yuchong Sun, Ruizhe Chen, Zhaohai Li, Chenxu L\"u, Zhibo Yang, Tao Yu, Xionghui Chen
arXiv:2602. 19710v3 Announce Type: replace-cross Abstract: Existing Vision-Language-Action (VLA) models often suffer from feature collapse and low training efficiency because they entangle high-level perception with sparse, embodiment-specific action supervision.
By Haitao Lin, Hanyang Yu, Jingshun Huang, He Zhang, Yonggen Ling, Ping Tan, Xiangyang Xue, Yanwei Fu