Xiaodong He

Career

Xiaodong He worked at Microsoft Research as a Principal Researcher and head of its Deep Learning Technology Center. After joining JD.com, Xiaodong He became a Senior Vice President of JD Group and managing director of JD AI Research, and the IEEE MMSP 2025 programme also lists Xiaodong He as managing director of the JD Joy Future Academy.¹

Xiaodong He is also an affiliate professor in the University of Washington's Department of Electrical and Computer Engineering. Xiaodong He gave a keynote titled "From Digital Human to Humanoids" at IEEE MMSP 2025, on building digital humans and humanoid systems with large language and vision models.²¹

Research

Xiaodong He's research spans deep learning, speech, natural language, computer vision and information retrieval. Vision-and-language work includes the 2017 bottom-up and top-down attention method for image captioning and visual question answering, and AttnGAN, an attentional generative adversarial network for fine-grained text-to-image generation.²⁴⁵

Work and contributions

• Deep Structured Semantic Models (DSSM) Published research

  Co-author of "Learning Deep Structured Semantic Models for Web Search using Clickthrough Data" (CIKM 2013).³

• Bottom-Up and Top-Down Attention Published research

  Co-author of "Bottom-Up and Top-Down Attention for Image Captioning and Visual Question Answering" (CVPR 2018), which proposes combining object-level bottom-up attention with top-down attention for image captioning and visual question answering.⁴

• AttnGAN Published research

  Co-author of the 2017 paper proposing AttnGAN, an attentional generative adversarial network for fine-grained text-to-image generation.⁵

Timeline

1. Jul 2017Co-authored the bottom-up and top-down attention paper for image captioning and visual question answering.⁴
2. Nov 2017Co-authored AttnGAN, an attentional GAN for text-to-image generation.⁵
3. Sep 21, 2025Gave the keynote "From Digital Human to Humanoids" at IEEE MMSP 2025.¹

Other documented roles

Roles the sources document that are not confirmed as current.

• Principal Researcher and Head of the Deep Learning Technology Center (DLTC), Microsoft Research¹²

Honors and recognition

• IEEE Fellow · Institute of Electrical and Electronics Engineers¹
• IEEE SPS Best Paper Award · IEEE Signal Processing Society¹

Where they work

• JD Group 1 person in the wiki
• University of Washington 19 people in the wiki

Related people

• Hassan ShapourianMultimodal models · Computer vision
• Jyoti AnejaMultimodal models · Computer vision
• Ali FarhadiComputer vision · Multimodal models
• Matt DeitkeMultimodal models · Computer vision
• Sahil DuaLanguage models · Multimodal models
• Nick FrosstLanguage models · Computer vision
• Hoifung PoonMultimodal models · Language models
• Nikhila RaviComputer vision · Multimodal models
• Sathwik Tejaswi MadhusudhanLanguage models · Multimodal models
• Jianfeng GaoLanguage models · Multimodal models