PeopleofSuperintelligence

Researcher

Gongshen Liu

Also known as 刘功申, Liu Gongshen

Gongshen Liu is an AI security researcher at Shanghai Jiao Tong University whose work spans natural language processing and the security of large models. He earned computer science degrees at Shandong University of Technology, Shandong University and SJTU, then taught at SJTU and visited Arizona State University. His research includes a 2005–2007 internet-text supervision project, a 2018 syntax-aware semantic-role-labeling paper, and later studies of backdoors and safety in language models. As of October 2026, he is a tenured professor at SJTU's Institute of Artificial Intelligence and Cognitive Security.

Works
237
Citations
2,391
h-index
23
Badges

Publication numbers from OpenAlex, as of 22 October 2026.

Timeline

  1. 1993–2003

    Computer science degrees

    He earns a B.A. in computer science at Shandong University of Technology in 1997, an M.A. in computer science at Shandong University in 2000, and a Ph.D. in computer science at Shanghai Jiao Tong University in 2003.

  2. 2004–2014

    Teaching at SJTU and ASU

    He joins SJTU as a lecturer in 2004, becomes an associate professor in 2008, and visits Arizona State University as a visiting scholar in 2014.

  3. 2005–2007

    Internet-text supervision

    At SJTU, he works on internet-text supervision using concept analysis in a National Natural Sciences Foundation project from 2005 to 2007.

  4. 2018

    Syntax-aware semantic role labeling

    In 2018, he coauthors “A Unified Syntax-aware Framework for Semantic Role Labeling,” which tests ways to incorporate syntax into neural networks for semantic role labeling.

  5. 2023

    A review of language-model backdoors

    In 2023, he coauthors a review of backdoor attacks and countermeasures in natural-language-processing models, organizing attacks by attacker capabilities and the affected stages of language-model development.

  6. 2024

    R-Judge and translation evaluation

    In 2024, he coauthors R-Judge, a benchmark for evaluating whether language models can identify safety risks in agent interaction records. In March 2024, his Artificial Intelligence Security Laboratory team provides technical support for an intelligent evaluation system at SJTU's 4th Siyuan Huawei Cup Creative Translation Competition.

  7. 2025

    The ALIS instruction security method

    In 2025, he coauthors ALIS, an instruction-tuning method that decomposes user input into atomic instructions and helps language models resist prompt-injection and jailbreaking attacks.

  8. As of October 2026

    Tenured professor at SJTU

    He is a tenured professor at SJTU's Institute of Artificial Intelligence and Cognitive Security. From August 2025 to August 2026, he leads a research project on power-grid cybersecurity and large-model applications. In 2026, he coauthors an ACL Findings paper on confidence calibration in diffusion language models and mathematical-reasoning benchmarks.

Career

  1. Since 2004Shanghai Jiao Tong UniversityLecturer, Associate professor
  2. 2014Arizona State UniversityVisiting scholar

Education

  1. 2003Shanghai Jiao Tong UniversityPh.D.
  2. 2000Shandong UniversityM.A.
  3. 1997Shandong University of TechnologyB.A.

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

  1. 84% of people shaping AI with recorded undergraduate and doctoral schools changed institutions for their doctorateIncludes Shanghai Jiao Tong University. 84% of people shaping AI with recorded undergraduate and doctoral schools changed institutions for their doctorate. Across 2,116 people with both an undergraduate degree and a doctorate recorded, we trace which undergraduate schools feed which doctoral programs, how often students stay at the same school, and how paths differ by birth-country region.
  2. The median h-index is 48 at universities and 14 at frontier AI labsIncludes Shanghai Jiao Tong University. 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.
  3. More people shaping AI studied at Massachusetts Institute of Technology than at any other institutionIncludes Shanghai Jiao Tong University. We 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.
  4. The company share of AI researchers’ top papers was 31% in 2020–2024 and 8% in 1970–1974Includes Shanghai Jiao Tong University. We estimate where the people shaping AI worked when their most-cited papers appeared: the company share was 31% in 2020–2024 and 8% in 1970–1974, and Google led all organizations in 2020–2024.

Colleagues

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

Sources and credits

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

  1. 上海交通大学-电子信息与电气工程学院-English tw.seiee.sjtu.edu.cn. Checked 21 October 2026.
  2. Gongshen Liu - ACL Anthology aclanthology.org. Checked 21 October 2026.
  3. [2309.06055] Backdoor Attacks and Countermeasures in Natural Language Processing Models: A Comprehensive Security Review arxiv.org. Checked 21 October 2026.
  4. [2401.10019] R-Judge: Benchmarking Safety Risk Awareness for LLM Agents arxiv.org. Checked 21 October 2026.
  5. The Award Ceremony of the 4th Siyuan Huawei Cup Creative Translation Competition Held at Shanghai Jiao Tong University - NEWS - 上海交通大学外国语学院 sfl.sjtu.edu.cn. Checked 21 October 2026.
  6. ALIS: Aligned LLM Instruction Security Strategy for Unsafe Input Prompt aclanthology.org. Checked 21 October 2026.
  7. 上海交大研究生院主页平台管理系统 Liu Gongshen--Home-- 电力网安垂城大模型及关键应用研究 faculty.sjtu.edu.cn. Checked 21 October 2026.
  8. Gongshen Liu-上海交通大学计算机学院（网络空间安全学院、密码学院） cs.sjtu.edu.cn. Checked 21 October 2026.