3 papers
cs.RO2026
Acting Through Registers
Yiwei Zhang, Yantai Yang, Jin Gao +7
World Action Models (WAMs) use generative visual representations to condition robot actions, yet how these representations organize action-relevant information remains poorly under…
cs.RO2026
LM-X: Explainable Vision--Language--Action Modeling via Progress, Event, and Uncertainty Prediction
Jin Lou, Zhiyuan Jing, Xupeng Wang +21
Large-scale vision--language--action (VLA) policies have advanced generalist robot control, yet most remain stimulus-to-action black boxes: actions are exposed, but their explanato…
cs.LG2026
SinkRouter: Sink-Aware Routing for Efficient Long-Context Decoding in Large Language and Multimodal Models
Junnan Liu, Xinyan Liu, Peifeng Gao +4
In long-context decoding for LLMs and LMMs, attention becomes increasingly memory-bound because each decoding step must load a large amount of KV-cache data from GPU memory. Existi…