Haolin Yang

Spatial reasoning in VLMs operates on coarse object localization rather than precise boundaries, revealing a surprising disconnect between knowing where objects are and how they relate.

Compact, automatically selected anatomical regions can drastically reduce hallucinations in medical VLMs without requiring expert annotations.

A novel data synthesis approach enables a lightweight model to outperform state-of-the-art planning systems by over 10x in household appliance manipulation tasks.

Transformers generalize out-of-distribution not by clever interpolation, but by learning a separate, orthogonal representation subspace for unseen tasks.