cv

Curriculum Vitae of Jea (Jay) Kwon

General Information

Name Jea (Jay) Kwon
Position Postdoctoral Researcher
Affiliation Max Planck Institute for Security and Privacy (MPI-SP), Bochum, Germany
Email onlytojay@gmail.com

Research Interest

  • Trustworthy AI: AI Interpretability [1, 2], AI Steerability [3, 4], AI Alignment [5, 6]
  • NeuroAI: Neuroscience for AI [7, 8], AI for Neuroscience [9, 10]
  1. AI Engram: In Search of Memory Traces in Artificial Intelligence, ICML 2026 Oral, Kwon et al. (First)
  2. Moir: Let the Model Direct Its Own Story for Robust Cross-Domain Knowledge Editing, ICML 2026 Mech. Interp. Workshop, Kwon et al. (First)
  3. Bilinear Relational Structure Fixes Reversal Curse and Enables Consistent Model Editing, ICLR 2026, Kim, Kwon et al. (Co-author)
  4. Erase or Hide? Suppressing Spurious Unlearning Neurons for Robust Unlearning, ICLR 2026, Yang, Kwon et al. (Co-author)
  5. Dropouts in Confidence: Moral Uncertainty in Human-LLM Alignment, AAAI 2026, Kwon et al. (First)
  6. Machine Behavior in Relational Moral Dilemmas: Moral Rightness, Predicted Human Behavior, and Model Decisions, ACL 2026 Findings, Kim, Kwon et al. (Co-corresponding)
  7. Transformer as a Hippocampal Memory Consolidation Model Based on NMDAR-Inspired Nonlinearity, NeurIPS 2023, Kim, Kwon et al. (First)
  8. Brain-inspired Lp-Convolution Benefits Large Kernels and Aligns Better with Visual Cortex, ICLR 2025, Kwon et al. (First)
  9. SUBTLE: An Unsupervised Platform with Temporal Link Embedding that Maps Animal Behavior, IJCV 2024, Kwon et al. (First)
  10. Cerebellar Tonic Inhibition Orchestrates the Maturation of Information Processing and Motor Coordination, Exp. Mol. Med. 2026, Kwon et al. (First)

Professional Experience

  • 2024 - Present
    Postdoctoral Researcher in Trustworthy AI
    Max Planck Institute for Security and Privacy (MPI-SP), Bochum, Germany
    • Advisor: Meeyoung Cha, Scientific Director of MPI-SP
  • 2022 - 2024
    Postdoctoral Researcher in NeuroAI
    Institute for Basic Science (IBS), Daejeon, South Korea
  • 2018 - 2020
    Researcher in Neuroscience
    Institute for Basic Science (IBS), Daejeon, South Korea
  • 2014 - 2018
    Student Researcher
    Korea Institute of Science and Technology (KIST), Seoul, South Korea

Education

  • 2014 - 2022
    Ph.D. in Nano-Bio-Information-Technology
    Korea University, Seoul, South Korea
    • Advisor: C. Justin Lee, Director of IBS
    • Full-Ride Scholarship, KU-KIST Graduate School
  • 2008 - 2012
    B.E. in Electrical and Electronic System
    Saitama University, Saitama, Japan
    • Full-Ride Scholarship, Korea-Japan Governments Joint Scholarship Program

Achievements

  • 2026
    Seal of Excellence — MSCA Postdoctoral Fellowships 2025
    European Commission, Horizon Europe (HORIZON-MSCA-2025-PF-01-01)
    • Proposal 101286042 — TRACE: Towards Reliable AI through Cognitive Engrams (with Max-Planck-Gesellschaft), total score 89.40%
  • 2023
    Excellent Paper Award
    Korean Artificial Intelligence Association (JKAIA)
    • Brain-inspired Lp-Convolution benefits large kernels and aligns better with visual cortex
  • 2023
    Patent Application & Technology Transfer
    Application Number: 10-2023-0027219
    • Device and Method for Classification of Human and Animal Behavior, Patent Contribution=35%
    • Finalized a technology transfer with ACTNOVA Inc.
  • 2022
    Best Presentation Award
    Korean Society for Brain and Neural Sciences (KSBNS)
    • ABCD-analysis: Mapping animal behavior and differential analysis of kinematic features without selection bias
  • 2022
    Young Investigator Award
    Korean Society for Molecular and Cellular Biology (KSMCB)
    • Retina-attached slice recording reveals light-triggered tonic GABA signaling in suprachiasmatic nucleus
  • 2021
    Best Poster Award
    Center for Cognition and Sociality (CCS) Workshop
    • Mapping Behavior with DECLARE: Deep Embedded Clustering of Action REpresentation

Talks

  • 2026
    AI Engram: In Search of Memory Traces in Artificial Neural Networks
    Scalable Trustworthy AI (STAI) group - KAIST
    • (Invited Seminar, In person, invited by Seong Joon Oh)
  • 2026
    AI Engram: In Search of Memory Traces in Artificial Neural Networks
    Max Planck Institute for Software Systems
    • (Invited Seminar, In person, invited by Krishna Gummadi)
  • 2026
    AI Engram: In Search of Memory Traces in Artificial Neural Networks
    Spark(l)ing Science - Max Planck Institute for Security and Privacy
    • (Seminar, In person)
  • 2025
    How neuroscience can benefit AI
    KAIST BK21 IBS Symposium - Integrated Neuroscience and Physiology
    • (Invited Seminar, In person, invited by Greg Suh)
  • 2025
    Transformer as a Hippocampal Memory Consolidation Model based on NMDAR-inspired Nonlinearity
    KSBNS Symposium
    • (Invited Seminar, In person)
  • 2025
    Sovereign AI: K-AI as an opportunity
    Netzwerk Junge Generation Deutschland-Korea
    • (Public Lecture, In person, invited by Muhong Lee)
  • 2025
    Brain-inspired AI
    Institute for Basic Science (IBS)
    • (Invited Seminar, In person, invited by C. Justin Lee)
  • 2025
    Brain-inspired AI
    Korea Brain Research Institute (KBRI)
    • (Invited Seminar, In person, invited by Jeongyeon Kim)
  • 2025
    Brain-inspired AI
    Gwangju Institute of Science and Technology (GIST)
    • (Invited Seminar, In person, invited by Sundong Kim)
  • 2025
    Neuroscience for Artificial Intelligence
    MLAI group, Yonsei University
    • (Invited Lecture, Offline, invited by Kyungwoo Song)
  • 2024
    Brain-inspired AI
    Kyungpook National University (KNU)
    • (Invited Seminar, In person, invited by Saekwang Nam)
  • 2024
    SUBTLE: An Unsupervised Platform with Temporal Link Embedding that Maps Animal Behavior
    Korea Institute of Ocean Science & Technology (KIOST)
    • (Invited Seminar, In person, invited by Soomee Kim)
  • 2024
    SUBTLE: An Unsupervised Platform with Temporal Link Embedding that Maps Animal Behavior
    Korea Brain Research Institute (KBRI)
    • (Invited Seminar, In person, invited by Jeongyeon Kim)
  • 2023
    Transformer as a hippocampal memory consolidation model based on NMDAR-inspired nonlinearity
    IBS-CCS Symposium
    • (Oral Presentation, In person)
  • 2023
    Memory editing in the brain - Eternal Sunshine of the Spotless Mind
    IBS Science in the Cinema Roof
    • (Public Lecture, In person)
  • 2023
    From human mind to artificial intelligence - Is ChatGPT a bridge
    Reading Festival at Jeonil High School
    • (Public Lecture, In person)
  • 2022
    Resolving clustering bias during animal behavior mapping and kinematic profiling
    IBS Data Science Group seminar
    • (Invited Seminar, Online, invited by Meeyoung Cha)
  • 2022
    Resemblances Between Transformer's Nonlinearity and NMDA Receptor Dynamics
    IBS Winter School on AI-Boosted Basic Science
    • (Invited Lecture, In person)
  • 2021
    Age-dependent motor coordination analysis by 3D pose-estimation using AVATAR
    IBS 2021 2nd Workshop
    • (Oral Presentation, In person)
  • 2020
    Backpropagation and the Brain
    KAIST-NeuroAI Journal Club
    • (Oral Presentation, Hybrid)
  • 2020
    Deep learning with NMDA receptor inspired activation function
    IBS 2019 Year-End Workshop
    • (Oral Presentation, In person)
  • 2019
    Development of a Low-cost, Comprehensive Recording System for Circadian Rhythm Behavior
    Korean Society for Molecular and Cellular Biology (KSMCB)
    • (Oral Presentation, In person)

Teaching

  • Guest Lecturer
    Georgia Tech CSE 8803 BMI "Brain-inspired Machine Intelligence", Fall 2024
  • Guest Lecturer
    Ewha Womans University G 18067 "Introduction to System Health", Spring 2026
  • Mentoring
    SoHyung Kim (Korea University, Undergraduate Intern), IBS 2020, 6 months
    • Outputs: CVPR2023-w, IJCV2024 (co-author)
  • Mentoring
    Hyewon Kim (Eulji University School of Medicine, Undergraduate Intern), IBS 2024, 9 months
    • Outputs: ICLR2024-w (co-author), Exp.Neurobiol.2024 (co-first author)
  • Mentoring
    Minsung Kim (Seoul National University, Ph.D. Intern, with Dong-Kyum Kim), MPI-SP 2024, 3 months
    • Outputs: ICLR2025-b, ACL2026 (first author)
  • Mentoring
    Jiseon Kim (KAIST, Ph.D. Intern, with Luiz Felipe Vecchietti), MPI-SP 2025, 6 months
    • Outputs: ICLR2025-w, ACL2026-f (first author)
  • Mentoring
    Nakyeong Yang (Seoul National University, Ph.D. Intern, with Dong-Kyum Kim), MPI-SP 2025, 3 months
    • Outputs: ICLR2026 (first author)

Community Service

  • Symposium Organizer
    KSBNS 2026 Annual Meeting
    • "Minds Meet Machines: Bidirectional Innovations in AI and Neuroscience"
  • Symposium Co-Organizer
    KSBNS & CJK 2025 Joint Symposium
    • "Neuroscience-inspired AI: Computational Insights into Biological and Artificial Intelligence"
  • Program Committee
    AAAI 2026, AI Alignment Track
  • Program Committee
    Machine+Behavior 2026, Harnack House, Berlin
  • Emergency Reviewer
    ICML 2026
  • Reviewer
    NeurIPS 2026
  • Reviewer
    ICML 2026 Mechanistic Interpretability Workshop
  • Reviewer
    ICLR 2026
  • Reviewer
    ICLR 2026 Workshop on Representational Alignment (Re-Align)
  • Reviewer
    ICLR 2025 Workshop on Representational Alignment (Re-Align)
  • Founding Member
    KAIST-NeuroAI Journal Club, 2020