Jingpu Yang

Multimodal LLMs can cleanly disentangle camera ego-motion from real-world scene dynamics without any drone telemetry or pose sensors, using just four optical-flow-derived residual tokens per slice to unlock robust aerial video reasoning.

LLM judges are swayed by style, mistaking it for substance, but a new framework can significantly mitigate this bias.

VLMs may ace dish recognition but often falter in delivering safe dietary advice, revealing a critical gap in their practical application for health management.

A novel zero-parameter gating mechanism boosts UAV video segmentation performance by over 4% while maintaining temporal stability.

Tactile feedback, when strategically sampled and evaluated, unlocks robust and safe robotic insertion policies even under sub-millimeter tolerances.