Hang Hua

Video world models can now maintain persistent scene memory without ever-growing storage by warping and fusing low-rank features into a fixed-size HexMemory.

Heterogeneous visual agents can boost downstream performance by up to 4.9 points across diverse tasks simply by reading from a shared, self-organizing 2B memory tree—without updating a single host model weight.

Memory consistency in video generation models falters significantly when objects disappear, with state-of-the-art models struggling to recover updated states upon reappearance.

LLMs show significant variability in the actionability of their UX critiques, with some models outperforming others across different product categories.

LLMs can now interpret quantum operations, achieving state-of-the-art results in circuit synthesis while allowing natural language constraints.

Success in long-horizon tasks hinges more on an agent's iterative persistence than on the quality of its initial solution.

Visual skills outperform text-only methods, revealing a critical gap in how agents learn and apply knowledge in complex, visual environments.

Soft-gating with an "advisor" model can steer LLMs to be safer and more useful, reducing over-refusal without sacrificing detection accuracy.

VLMs can now get a million-scale boost in chart-understanding abilities thanks to a new dataset with paired code, images, data, and reasoning.