Sibo Liu

A new hierarchical framework for image style reveals an expansive style space, enhancing both diversity and semantic richness in visual representation.

A single authorization rule can effectively prevent memory leakage across different audience groups in language agents, ensuring that sensitive information remains compartmentalized.

Style transfer just got a whole lot easier: MegaStyle's 1.4M image dataset, generated via text-to-image models, unlocks reliable style similarity measurement and generalizable style transfer.