PeopleofSuperintelligence

Researcher

Bochao Wu

Bochao Wu works in DeepSeek's Research & Engineering group and coauthors research on its large language models. DeepSeek-V3's 2024 report describes a 671-billion-parameter mixture-of-experts model and multi-token prediction. DeepSeek-R1 applies reinforcement learning to language-model reasoning, while V3.2 adds sparse attention and a scalable reinforcement-learning framework. In 2026, DeepSeek-V4 supports million-token contexts, and V4.1-Flash pairs KV-cache reuse with FP4 caching.

Now at
DeepSeek
Works
40
Citations
2,267
h-index
13
Badges

Publication numbers from OpenAlex, as of 19 October 2026.

Timeline

  1. 2024

    DeepSeek-V3's training design

    He coauthors the DeepSeek-V3 technical report in 2024, which describes a 671-billion-parameter mixture-of-experts model and a multi-token prediction objective.

  2. 2025

    Reasoning by reinforcement learning

    In 2025, he coauthors DeepSeek-R1, whose Nature paper shows how pure reinforcement learning can develop self-reflection and verification in language models.

  3. December 2025

    Sparse attention in V3.2

    He coauthors DeepSeek-V3.2 in December 2025; the model introduces DeepSeek Sparse Attention and a scalable reinforcement-learning framework for reasoning and agent performance.

  4. April 2026

    Million-token DeepSeek-V4

    DeepSeek-V4 names him as a coauthor in April 2026 and presents two mixture-of-experts models that support one million tokens of context.

  5. As of October 2026

    Research and engineering at DeepSeek

    As of October 2026, he works in DeepSeek's Research & Engineering group. The 2026 DeepSeek-V4.1-Flash model pairs CSA2 cache reuse with FP4 caching in a multimodal system for one-million-token contexts.

Career

  1. Since 2025DeepSeek
  2. 2024–2026Northwestern Polytechnical University
  3. 2024–2026Yuanpei University
  4. 2019–2022Chinese Academy of Sciences
  5. 2019–2022University of Chinese Academy of Sciences
  6. 2018–2024Nantong University

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

Most-cited works

  1. DeepSeek-R1 incentivizes reasoning in LLMs through reinforcement learning2025 · Nature · 1,112 citationsWith Bingxuan Wang, Chenggang Zhao, Daya Guo and Zhibin Gou
  2. Can Open Large Language Models Catch Vulnerabilities?2025 · 538 citationsWith Bingxuan Wang, Chenggang Zhao and Zhibin Gou
  3. DeepSeek-V3 Technical Report2024 · 292 citationsWith Bingxuan Wang, Chenggang Zhao and Daya Guo
  4. Transitions and Origin of the Type-B Quasi-periodic Oscillations in the Black Hole X-Ray Binary MAXI J1348–6302022 · The Astrophysical Journal · 31 citations
  5. Adaptive active fault-tolerant MPPT control for wind power generation system under partial loss of actuator effectiveness2018 · International Journal of Electrical Power & Energy Systems · 29 citations

From OpenAlex, as of 19 October 2026.

  1. The median h-index is 54 at universities and 19 at frontier AI labsIncludes Bochao Wu. The median h-index is 54 at universities and 19 at frontier AI labs. We compare publishing careers, citations and moves to estimate how universities and frontier labs contribute to AI research.
  2. The company share of AI researchers’ top papers rose from 2% to 32%Includes DeepSeek. We estimate where leading AI researchers worked when their most-cited papers appeared. The company share rose from 2% in 1975–1979 to 32% in 2020–2024.
  3. Google alumni founded or led 81 other organizations, the most among the labs we compareIncludes DeepSeek. We trace 194 organizations founded or led by AI lab alumni, and 202 career moves between the labs.
  4. Stanford University trained or employed 26 of the 108 AI major-prize winners, more than any other universityIncludes Nantong University. Stanford University trained or employed 26 of the 108 AI major-prize winners, more than any other university. We compare doctorate and employment links for major-prize and best-paper winners.

Colleagues

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

Sources and credits

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

  1. [2412.19437] DeepSeek-V3 Technical Report arxiv.org. Checked 18 October 2026.
  2. DeepSeek-V3 Technical Report | alphaXiv alphaxiv.org. Checked 18 October 2026.
  3. [2501.12948] DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning arxiv.org. Checked 18 October 2026.
  4. DeepSeek-R1 incentivizes reasoning in LLMs through reinforcement learning | Nature nature.com. Checked 18 October 2026.
  5. [2512.02556] DeepSeek-V3.2: Pushing the Frontier of Open Large Language Models arxiv.org. Checked 18 October 2026.
  6. [2606.19348] DeepSeek-V4: Towards Highly Efficient Million-Token Context Intelligence arxiv.org. Checked 18 October 2026.
  7. DeepSeek-V4.1-Flash: Pushing the Limits of KV Cache Compression arxiv.org. Checked 18 October 2026.
  8. [2609.19969] DeepSeek-V4.1-Flash: Pushing the Limits of KV Cache Compression arxiv.org. Checked 18 October 2026.