PeopleofSuperintelligence

Researcher

Zihan Qiu

Zihan Qiu is a researcher on Alibaba Group's Qwen Team whose work spans large language models, expert routing, and transformer attention. He studied in Tsinghua University's Yao Class before joining the team and coauthored technical reports on Qwen2.5 and Qwen3. A 2025 paper on gated attention won a NeurIPS Best Paper award. In 2026, he coauthored a report on Qwen3.8-Next and presented research on outliers in large language models.

Now at
Alibaba
Works
64
Citations
675
h-index
11
Badges

Publication numbers from OpenAlex, as of 18 October 2026.

Timeline

  1. 2022–2024

    Tsinghua to Qwen Team

    From 2022 to 2024, he studies in Tsinghua University's Yao Class and moves to Alibaba's Qwen Team; his interests include language models and modular deep learning.

  2. 2024

    The Qwen2.5 report

    In 2024, he coauthors the Qwen2.5 Technical Report, which introduces a series of large language models designed to meet diverse needs.

  3. 2025

    A recurrent expert router

    In 2025, he coauthors Layerwise Recurrent Router for Mixture-of-Experts, which uses a gated recurrent unit to connect routing decisions across consecutive layers.

  4. 2025

    The Qwen3 report

    In 2025, he coauthors Qwen3 Technical Report, which presents the latest Qwen model family with a series of large language models and multilingual capabilities.

  5. 2025

    A NeurIPS Best Paper

    In 2025, a paper coauthored by Qiu wins a NeurIPS Best Paper award for Gated Attention for Large Language Models. Its authors find that a head-specific sigmoid gate after scaled dot-product attention improves performance.

  6. 2026

    The Qwen3.8-Next report

    In 2026, he coauthors a report on the Qwen3.8-Next architecture. It describes Qwen3.8-Flash-Next as a sparse mixture-of-experts model with 125 billion parameters.

  7. As of October 2026

    Researcher at Qwen

    As of October 2026, he is on Alibaba Group's Qwen Team. In 2026, he presents work on outliers in large language models.

A language model matched the links in this timeline to people, organizations and papers (provisional, model-judged).

Career

  1. Since 2024AlibabaResearcher
  2. 2023–2026Xi'an Jiaotong University
  3. 2019–2023Kent State University

Education

  1. 2024Tsinghua University

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

Most-cited works

  1. Qwen3 Technical Report2025 · 147 citationsWith Fei Huang, Le Yu and Rui Men
  2. Qwen2.5 Technical Report2024 · 122 citationsWith Fei Huang, Le Yu and Rui Men
  3. A survey on multilingual large language models: corpora, alignment, and bias2025 · Frontiers of Computer Science · 71 citations
  4. Fatty acid-like Pt( iv ) prodrugs overcome cisplatin resistance in ovarian cancer by harnessing CD362020 · Chemical Communications · 46 citations
  5. A combined predicting model for benign esophageal stenosis after simultaneous integrated boost in esophageal squamous cell carcinoma patients (GASTO1072)2022 · Frontiers in Oncology · 38 citations

From OpenAlex, as of 18 October 2026.

  1. Google alumni founded or led 77 other organizations, the most among the labs we compareIncludes Alibaba. We trace 177 organizations founded or led by AI lab alumni, and 186 career moves between the labs.
  2. The company share of AI researchers’ top papers rose from 2% to 33%Includes Alibaba. We estimate where leading AI researchers worked when their most-cited papers appeared. The company share rose from 2% in 1975–1979 to 33% in 2020–2024.
  3. Google has the highest combined h-index among AI organizationsIncludes Alibaba. Our estimates of combined and median h-index at AI organizations, using current affiliations and OpenAlex records.
  4. Stanford University trained or employed 27 of the 113 AI major-prize winners, more than any other universityIncludes Tsinghua University. Stanford University trained or employed 27 of the 113 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 Zihan Qiu yet. Send us a free-licensed photo.

  1. Big AI Dream | Jie Fu bigaidream.github.io. Checked 18 October 2026.
  2. [2412.15115] Qwen2.5 Technical Report arxiv.org. Checked 18 October 2026.
  3. Layerwise Recurrent Router for Mixture-of-Experts proceedings.iclr.cc. Checked 18 October 2026.
  4. [2505.09388] Qwen3 Technical Report arxiv.org. Checked 18 October 2026.
  5. NeurIPS 2025 Awards neurips.cc. Checked 18 October 2026.
  6. Announcing the NeurIPS 2025 Best Paper Awards – NeurIPS Blog blog.neurips.cc. Checked 18 October 2026.
  7. [2608.30320] On the Design of Qwen3.8-Next Architecture: Evaluation, Efficiency, and Training Stability arxiv.org. Checked 18 October 2026.
  8. Workshop on High-dimensional Learning Dynamics - Schedule sites.google.com. Checked 18 October 2026.