Xiaohua Zhai

Research Scientist at Meta

Zhai helped create widely used vision-language resources including WebLI, SigLIP and PaliGemma, improving how models connect images and text and making strong multimodal systems available as open weights. She now researches multimodal AI at Meta.