Recent embodied systems increasingly build on multimodal large language models (MLLMs) for their pretrained understanding and reasoning capabilities, but their perception of the physical world remains largely limited to basic modalities such as vision and audio. However, broadening MLLMs to more diverse sensory modalit...
Shi-Yun Xiong, Dong-Ming Wu, Pei-Wen Sun et al.· 0 citations
Foundation models are driving a paradigm shift in robotic intelligence, moving beyond low-level action prediction to goal-directed orchestration of data collection, model training, task planning, and real-world execution. We characterize this emerging paradigm as an agentic robots, in which agents reason over human ins...
Chonghan Qin, Dong-Ming Wu, Kai-Tuo Feng et al.· 2 citations
Learning-based manipulation policies relying on RGB cameras often suffer from degraded observations under extreme exposure. Event cameras mitigate this degradation by asynchronously detecting pixel-level intensity changes to offer a high dynamic range. However, their observations heavily depend on camera placement, as...
Xin-Yue Wang, Yi-Cheng Jiang, Ze-Sen Gan et al.· 0 citations
Instruction following is central to language-conditioned robot policies: language should determine what to do when the same scene permits multiple valid actions. Yet successful execution alone cannot establish whether a policy follows the instruction or infers the task from the scene. We study this ambiguity through sc...
Shao-Han Jiang, Jia-Hang Cao, Qi-Duo He et al.· 1 citation
Most vision-language-action models represent future motion as fixed-rate action chunks, tying temporal resolution and prediction horizon to a fixed output budget. This pointwise representation wastes capacity on highly correlated neighboring actions, leaves temporal continuity and smoothness to be learned implicitly, a...
Yi-Fan Li, Jia-Xu Wang, Dong-Ming Wu et al.· 0 citations
AR-WAM is presented, a visual-conditioned, agent-ready world action model that replaces language with two complementary conditions: a visual grounding prompt and a learnable operation token dictating the atomic skill to execute, predicting scene evolution within compact latent states while decoding actions.
Yi-Cheng Jiang, Ze-Sen Gan, Xiao-Bo Wang et al.· arXiv.org· 3 citations
Across in-domain and distribution-shifted settings, allocating additional test-time computation substantially improves next-subtask prediction accuracy, and these gains translate into higher closed-loop success on long-horizon robot manipulation tasks.