Jianfei Yang

Robots can now achieve over double the success rate in using novel tools by leveraging keypoint trajectory reasoning for functional generalization.

ActiveVital achieves vital signs monitoring accuracy comparable to static methods by actively controlling radar alignment, reducing respiration interval error from 0.87 s to 0.14 s.

Spatial-semantic prompting outperforms traditional text-only methods in embodied visual tracking, especially in complex environments with similar distractors.

Humanoid robots can now achieve superior stability and accuracy in unpredictable environments by leveraging a physics-based disturbance observer that operates without force sensors.

APEX reduces tracking error by over 41% and boosts manipulation success rates significantly, all without altering existing policies or controllers.

Gesture understanding can elevate robotic manipulation success rates by 80%, transforming how robots interpret human intentions.

AerialClaw empowers UAVs to autonomously interpret and adapt to missions in real-time, revolutionizing aerial operations beyond static command sequences.

Robots can now understand and act on your gaze while following language commands, enabling more intuitive and precise human-robot collaboration.

Ditch brittle token rankings: OccamToken uses register-anchored relative evidence testing to prune visual tokens in VLMs, achieving extreme compression (down to 1.4%!) without retraining or significant accuracy loss.

Robots can learn faster and perform better by strategically choosing *when* to be unpredictable, achieving both multimodal expressivity and deterministic efficiency.

Robots can now better understand implicit instructions in cluttered scenes, choosing the knife over the scissors when told to "cut the apple," thanks to a new benchmark and neural network architecture designed to handle functionally similar objects.

LLMs can navigate complex 3D environments more effectively and with far fewer tokens by using a hierarchical scene graph representation derived from omnidirectional sensor data.

Forget retargeting: RoboForge's physics-optimized pipeline lets humanoids nail text-guided locomotion with better accuracy and stability.