Tag
RoboQuest introduces a benchmark for goal-directed embodied exploration in mobile manipulation, testing frontier multimodal agents on search, inspection, and interactive testing tasks. Results show the best agent succeeds in only 23% of episodes, with failures mainly attributed to premature stopping of exploration rather than execution skills.
MM-ABC提出一个面向通用移动操作的基础模型，通过稀疏多层VLM特征实现空间感知、以世界想象与几何意图为监督的训练分支，以及MM-APT中用掩码联合注意力协调操作与移动流。模型在5000+小时、17种本体的数据上预训练，在RoboCasa365、LIBERO等基准及真实移动操作任务上取得强性能。
The paper introduces DECOWAM, a decoupled whole-body world-action model for legged mobile manipulation that improves video and action prediction performance over existing models like FastWAM through dedicated conditional interfaces and a new dataset.
Xiaomi released XR-1, a robot foundation model trained on over 100K hours of real-world manipulation trajectories. Built on Qwen3-VL and a Diffusion Transformer, it enables out-of-the-box mobile manipulation in unseen environments.
Unitree releases UnifoLM-OminiA-0.3, a single AI model for whole-body mobile manipulation with real-time omni-modal interaction, enabling autonomous home-care and wellness tasks.
LimX Dynamics released COSA 0.5, an update to its humanoid brain system featuring a long-take demonstration of fully autonomous mobile manipulation without teleoperation, advancing humanoid robot autonomy.
ABot-M0.5 is a new World Action Model for mobile manipulation that improves performance through temporal granularity alignment, action space disentanglement, and train-test consistency, achieving state-of-the-art results on long-horizon and fine-grained manipulation benchmarks.
EBench is a diagnostic benchmark for generalist mobile manipulation policies, providing a multi-dimensional profile across 26 tasks and 4 generalization axes, revealing structural strengths and weaknesses beyond aggregate success rates.