PeopleofSuperintelligence

Researcher

Yiyang Ma

Also known as 马逸扬

Yiyang Ma is an AI researcher whose papers span image generation, joint audio-video systems, and DeepSeek's vision-language and reasoning models. At Peking University, they earned bachelor's and master's degrees; they also interned at Microsoft Research Asia. Their papers include AI Illustrator, MM-Diffusion, DeepSeek-VL2, DeepSeek-V3, JanusFlow and DeepSeek-R1. From June to November 2024, they worked as an AGI research intern at DeepSeek-AI. Since April 2025, they are an AGI researcher there.

Now at
DeepSeek
Works
13
Citations
436
h-index
6
Badges

Publication numbers from OpenAlex, as of 22 October 2026.

Timeline

  1. 2018–2025

    Peking University degrees

    From 2018 to 2022, they study intelligence science and technology at Peking University. From 2022 to 2025, they study data science there and earn a master's degree under Jiaying Liu.

  2. 2022

    Internship and AI Illustrator

    From January to August 2022, they work as a research intern in Microsoft Research Asia's Multimedia Search and Mining Group. In October 2022, they coauthor AI Illustrator, an ACM Multimedia paper that translates complex descriptions into images.

  3. 2023

    Joint audio-video generation

    In 2023, they coauthor MM-Diffusion, a framework for joint audio-video generation that uses two coupled denoising autoencoders.

  4. June–November 2024

    An AGI internship at DeepSeek-AI

    From June to November 2024, they work as an AGI research intern at DeepSeek-AI.

  5. December 2024

    DeepSeek-VL2 vision-language models

    In December 2024, they coauthor DeepSeek-VL2, a mixture-of-experts vision-language model series for high-resolution images, visual question answering, optical character recognition and visual grounding.

  6. December 2024

    The DeepSeek-V3 report

    In December 2024, they coauthor the DeepSeek-V3 Technical Report, which describes multi-head latent attention and DeepSeekMoE as tools for efficient inference and training.

  7. 2025

    JanusFlow and DeepSeek-R1

    In 2025, they coauthor JanusFlow, a framework that unifies image understanding and generation in one model. They also coauthor DeepSeek-R1 in 2025, which presents a pure reinforcement-learning approach to language-model reasoning without human-labeled reasoning trajectories.

  8. Since April 2025

    AGI researcher at DeepSeek-AI

    Since April 2025, they work as an AGI researcher at DeepSeek-AI.

Career

  1. Since 2024DeepSeekAGI Research Intern, AGI Researcher
  2. 2025–2026Guangzhou University
  3. 2025–2026Beijing National Laboratory for Molecular Sciences
  4. 2022Microsoft Research AsiaResearch Intern

Education

  1. 2025Peking UniversityMaster's degree
  2. 2022Peking UniversityBachelor's degree

From the biography and OpenAlex affiliations (the years of papers that list the organization), so some years are approximate.

  1. The median h-index is 48 at universities and 14 at frontier AI labsIncludes Yiyang Ma. The median h-index is 48 at universities and 14 at frontier AI labs. We compare publishing careers, recent citations, dual appointments and job moves across 2,420 university researchers and 308 frontier-lab researchers.
  2. 16% of AI career moves in the 2020s went from a university to a companyIncludes Yiyang Ma. We estimate that 16% of AI career moves in the 2020s went from a university to a company; 29% of moves starting at a university went to a company, and 76% ended the university role.
  3. Companies supply the most prior-work hires at 7 of the 9 frontier AI labsIncludes DeepSeek. Our estimates of the organizations where 550 people worked directly before joining 9 frontier AI labs across 620 career transitions, comparing the main feeder organizations per lab.
  4. 5,000 frontier users would be 38.2% of the core labs' 13,100 reported staffIncludes DeepSeek. 5,000 frontier users would be 38.2% of the core labs' 13,100 reported staff. We test the estimate against 3,913 named model contributors and dated all-staff records across Google DeepMind, OpenAI and Anthropic, plus six additional frontier-grade labs.

Colleagues

People with dated positions at the same organization in the same years.

Sources and credits

Photo: We have no published photo of Yiyang Ma yet. Send us a free-licensed photo.

  1. Ma, Yiyang | AGI Researcher @ DeepSeek-AI realpasu.github.io. Checked 20 October 2026.
  2. AI Illustrator: Translating Raw Descriptions into Images by Prompt-based Cross-Modal Generation microsoft.com. Checked 20 October 2026.
  3. MM-Diffusion: Learning Multi-Modal Diffusion Models for Joint Audio and Video Generation arxiv.org. Checked 20 October 2026.
  4. DeepSeek-VL2: Mixture-of-Experts Vision-Language Models for Advanced Multimodal Understanding arxiv.org. Checked 20 October 2026.
  5. DeepSeek-V3 Technical Report arxiv.org. Checked 20 October 2026.
  6. JanusFlow: Harmonizing Autoregression and Rectified Flow for Unified Multimodal Understanding and Generation openaccess.thecvf.com. Checked 20 October 2026.
  7. DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning arxiv.org. Checked 20 October 2026.