Agentic robotics systems combine planners with robot harnesses to perform complex manipulation tasks. Recent work improves these systems through better harnesses and external memory, but relies on powerful closed-source models as planners and does not improve the planners'own capabilities. In contrast, coding agents ha...
Xin-Xin Liu, Bing-Xin Xu, Wen-Rui Bao et al.· 1 citation· ⚡1
This work evaluates coding agents under a safety constraint, where each task pairs a manipulation goal with an obstacle the robot must not touch, and presents SafeHarness, which equips the model with two obstacle-aware harnesses.
Bing-Xin Xu, Yu-Zhang Shang, Zhen-Han Dong et al.· 2 citations
Agentic robotics places a large language model (LLM) agent in the robot control loop, where it composes perception, motion, and vision-language-action (VLA) tools under the guidance of a memory and thereby generalizes to new tasks without retraining. This generality comes at a steep price in efficiency. Every primitive...
Bing-Xin Xu, Wen-Rui Bao, Xin-Xin Liu et al.· 0 citations
LLM agents that orchestrate frozen vision-language-action (VLA) policies improve across episodes through text memory, which records what the agent did but not how the task is done. A demonstration video shows it, but fits poorly into an agent's context. The full video slows every turn, fixed keyframes lose the contact...
Wen-Rui Bao, Xin-Xin Liu, Bing-Xin Xu et al.· 1 citation
This work introduces DeltaWAM, a change-centric WAM that makes a compact delta token the unit of future prediction, each token is a single vector encoding changes between consecutive dense DINO feature maps.
Tian-Yun Jiang, Wen-Rui Bao, Bing-Xin Xu et al.· 0 citations
Coding agents have emerged as a promising paradigm for robot manipulation: a language model writes the robot controller as a program, and agents built in this way now operate robots without robot-specific training.Whether this paradigm is also safe, however, has not been asked. We evaluate coding agent under a safety c...
BATON improves task success by 11.6% and cumulative success by 14.9% over the SoTA and improves task success by 11.6% over the RoboMemArena on the long-horizon benchmark RoboMemArena.