PeopleofSuperintelligence

Researcher

Yiliang Xiong

Yiliang Xiong is a research author on DeepSeek's language models, with work spanning open-source LLM scaling, Mixture-of-Experts systems, sparse attention and reinforcement-learning reasoning. He coauthored DeepSeek LLM in January 2024, DeepSeek-V2 and V3 in 2024, DeepSeek-R1 and V3.2 in 2025, and DeepSeek-V4 in April 2026. DeepSeek-V3.2-Speciale reaches gold-medal performance at the 2025 International Mathematical Olympiad and International Olympiad in Informatics. As of October 2026, he is a co-author of the DeepSeek-AI report on DeepSeek-V4.1-Flash.

Timeline

  1. January 2024

    DeepSeek LLM scaling study

    In January 2024, he coauthors DeepSeek LLM, a project on open-source scaling whose report describes 7B and 67B model configurations and a two-trillion-token pretraining dataset.

  2. May 2024

    DeepSeek-V2's MoE design

    In May 2024, he coauthors DeepSeek-V2, a Mixture-of-Experts language model with 236B total parameters, 21B activated for each token and a 128K-token context window.

  3. December 2024

    DeepSeek-V3 Technical Report

    In December 2024, he coauthors the DeepSeek-V3 Technical Report, which describes a Mixture-of-Experts language model with 671B total parameters and 37B activated for each token.

  4. January 2025–September 2025

    Reasoning with DeepSeek-R1

    In January 2025, he coauthors DeepSeek-R1, which shows that pure reinforcement learning can incentivize language-model reasoning without human-labeled reasoning trajectories. Nature publishes the paper in September 2025, listing him among DeepSeek-AI Team authors in Hangzhou, China.

  5. December 2025

    Sparse attention in V3.2

    In December 2025, he coauthors DeepSeek-V3.2, which introduces DeepSeek Sparse Attention for long-context efficiency. Its high-compute variant, DeepSeek-V3.2-Speciale, reaches gold-medal performance at both the 2025 International Mathematical Olympiad and the International Olympiad in Informatics.

  6. April 2026

    DeepSeek-V4 preview

    In April 2026, he coauthors the DeepSeek-V4 preview, which presents two Mixture-of-Experts models: DeepSeek-V4-Pro with 1.6T parameters and DeepSeek-V4-Flash with 284B; both support one-million-token contexts.

  7. As of October 2026

    DeepSeek-V4.1-Flash report

    As of October 2026, he is a co-author of DeepSeek-AI's report on DeepSeek-V4.1-Flash, submitted in September 2026. The report introduces a multimodal Mixture-of-Experts model with 552B backbone parameters and contexts up to one million tokens.

  1. Google had the most people shaping AI every year from 2014 to 2025Our estimates of which organizations employed the most people shaping AI each year from 1955 to 2025, tracing the shift from university labs to companies.
  2. The United States employs an estimated 48% of people shaping AIWe estimate where the people shaping AI work today, how universities, companies and research institutes split across countries, and how country shares have changed since the 1960s.
  3. 43% of living people shaping AI work outside their country of birthWe estimate that 43% of living people shaping AI work outside their country of birth (520 of 1,198 people with known birth and employment countries).
  4. More people shaping AI studied at Massachusetts Institute of Technology than at any other institutionWe estimate where the people shaping AI studied and where their careers led. Massachusetts Institute of Technology leads with 467 former students, including 291 recorded doctorates.

Sources and credits

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

  1. DeepSeek LLM: Scaling Open-Source Language Models with Longtermism arxiv.org. Checked 21 October 2026.
  2. DeepSeek-V2: A Strong, Economical, and Efficient Mixture-of-Experts Language Model arxiv.org. Checked 21 October 2026.
  3. DeepSeek-V3 Technical Report arxiv.org. Checked 21 October 2026.
  4. DeepSeek-V3 Technical Report | European Laboratory for Learning and Intelligent Systems ellis.eu. Checked 21 October 2026.
  5. DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning arxiv.org. Checked 21 October 2026.
  6. DeepSeek-R1 incentivizes reasoning in LLMs through reinforcement learning | Nature nature.com. Checked 21 October 2026.
  7. DeepSeek-V3.2: Pushing the Frontier of Open Large Language Models arxiv.org. Checked 21 October 2026.
  8. DeepSeek-V4: Towards Highly Efficient Million-Token Context Intelligence arxiv.org. Checked 21 October 2026.
  9. DeepSeek-V4.1-Flash: Pushing the Limits of KV Cache Compression arxiv.org. Checked 21 October 2026.