PeopleofSuperintelligence

Researcher

Mintong Kang

Mintong Kang is a computer science Ph.D. candidate whose work develops AI safety and security methods, from fairness in machine learning to red-teaming of language models and agents. They co-first-authored DecodingTrust, a GPT-4 and GPT-3.5 trustworthiness study that received a 2023 NeurIPS Outstanding Paper Award. Later, they coauthored certified risk analysis for retrieval-augmented language models and a safety platform for multimodal models. As of October 2026, they are a Ph.D. candidate in computer science at the University of Illinois Urbana-Champaign and a Student Researcher at Google DeepMind.

Timeline

  1. 2021–2022

    Few-shot learning and fairness

    In 2021, they coauthor MgSvF, a TPAMI paper on few-shot class-incremental learning. Their 2022 NeurIPS Spotlight paper certifies distributional fairness by decomposing data into analytical subpopulations.

  2. 2023

    DecodingTrust and a NeurIPS award

    They co-first-author DecodingTrust, which evaluates GPT-4 and GPT-3.5 across toxicity, stereotype bias, adversarial robustness, privacy, machine ethics, and fairness. NeurIPS gives the 2023 paper an Outstanding Paper Award.

  3. 2024

    Certifying generation risks with C-RAG

    They coauthor C-RAG, a framework that certifies generation risks for retrieval-augmented language models. Its analysis gives theoretical guarantees under test-distribution shifts, and the paper is accepted to ICML 2024.

  4. 2025

    A multimodal safety benchmark

    They coauthor MMDT, a platform for evaluating safety and trustworthiness in multimodal foundation models. It measures safety, hallucination, fairness, privacy, adversarial robustness, and out-of-distribution generalization. The paper appears at ICLR 2025.

  5. As of October 2026

    Student researcher at Google DeepMind

    They are a Ph.D. candidate in computer science at UIUC, advised by Bo Li, and a Student Researcher at Google DeepMind. Their 2026 DTap paper presents a controllable red-teaming platform for agents across 14 real-world domains and more than 50 simulation environments.

Career

  1. NowGoogle DeepMindStudent Researcher

Education

  1. University of Illinois Urbana-ChampaignPh.D. candidate

From the biography.

  1. Google had the most people shaping AI every year from 2014 to 2025Includes Google. Our 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. Google has the highest combined h-index among AI organizationsIncludes Google. We estimate that Google holds the highest combined h-index at 8,016, while universities account for 73% of the deduplicated total across AI organizations.
  3. 20% of people at companies who shape AI also hold a university role, led by Google and Stanford UniversityIncludes Google. We estimate that 539 of the 6,766 people with a known current organization (8%) hold both a university role and a company role, including 20% of people at companies and 13% of people at universities.
  4. Google alumni have founded or led 133 other organizations, more than any other AI labIncludes Google. We estimate that alumni of 17 major AI labs went on to found or lead 350 other organizations, including 226 they founded, and we trace 472 career moves directly between 23 labs since 2010.

Colleagues

People named in this biography, and people with dated positions at the same organization in the same years.

Sources and credits

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

  1. Mintong Kang | Home kangmintong.github.io. Checked 21 October 2026.
  2. [2205.15494] Certifying Some Distributional Fairness with Subpopulation Decomposition arxiv.org. Checked 21 October 2026.
  3. NeurIPS 2023 Awards neurips.cc. Checked 21 October 2026.
  4. [2306.11698] DecodingTrust: A Comprehensive Assessment of Trustworthiness in GPT Models arxiv.org. Checked 21 October 2026.
  5. [2402.03181] C-RAG: Certified Generation Risks for Retrieval-Augmented Language Models arxiv.org. Checked 21 October 2026.
  6. [2503.14827] MMDT: Decoding the Trustworthiness and Safety of Multimodal Foundation Models arxiv.org. Checked 21 October 2026.
  7. [2605.04808] DecodingTrust-Agent Platform (DTap): A Controllable and Interactive Red-Teaming Platform for AI Agents arxiv.org. Checked 21 October 2026.