Jin Hyun Park ๋ฐ์งํ
Ph.D. student ยท Brain Networks Lab ยท Computer Science and Engineering, Texas A&M University
I try to teach neural networks a few tricks from real neurons.
I am a Ph.D. student at Texas A&M University, advised by Yoonsuck Choe. My research sits where machine learning meets computational neuroscience, with a side interest in natural language processing. During my Ph.D., I have worked on (1) biologically inspired neural networks, (2) the evolution of prediction, and (3) an AI-driven smart floor-heating thermostat. Since February 2026 I have also been a Physical AI Ph.D. Fellow with Doosan Corporation.
Before Texas A&M, I studied computer science at the University of Bristol (B.Sc., First Class Honours), where I was advised by Rui Ponte Costa and Cian O'Donnell. I have also worked at Hyundai Motor Company, LG Uplus (previously LG Telecom), Samsung Life Insurance, Tilda Corp., and the Multimodal Language and Cognition Lab at Kyungpook National University.
jinhyun.park@tamu.edu
CV /
Google Scholar /
GitHub /
LinkedIn
News
| Aug 2026 | KoNeoBench accepted to Findings of EMNLP 2026. |
| Feb 2026 | Joined Doosan Corporation as a Physical AI Ph.D. Fellow. |
| Jan 2026 | Attended CES 2026 at the invitation of Doosan Corporation. |
Publications
C: conference, J: journal, W: workshop, A: arXiv preprint. * denotes equal contribution.
| [C2] | Soha Lee, Soojin Lee, Heesung Yang, Hyunju Song, Hyunji Lee, Jinsan An, Jeongwan Shin, Jin Hyun Park, Jun Lee, Hyeyoung Park, Kilim Nam. KoNeoBench: A Curated Evaluation Dataset for LLM Understanding of Korean Neologisms. Findings of EMNLP, 2026. |
| [C1] | Seyyed Ali Ayati*, Jin Hyun Park*, Yichen Cai, Marcus Botacin. Making Acoustic Side-Channel Attacks on Noisy Keyboards Viable with LLM-Assisted Spectrograms' "Typo" Correction. 19th USENIX Workshop on Offensive Technologies (WOOT), 2025. [arXiv] |
| [J1] | William Kang, Chris Anand, Jin Hyun Park, Yoonsuck Choe. Evolutionary Factors Contributing to the Emergence of Prediction. Adaptive Behavior, 2025. [doi] |
| [W1] | Alex Alvarez*, Jin Hyun Park*, Yoonsuck Choe. Saliency Thresholds in Neural Code and its Relation to the Power-Law, Gaussian, and Lambert W Function. NeurIPS Workshop on Symmetry and Geometry in Neural Representations (NeurReps), 2025. [OpenReview] |
| [A2] | Jin Hyun Park, Cheng Zhang, Yoonsuck Choe. Incorporating Visual Cortical Lateral Connection Properties into CNN: Recurrent Activation and Excitatory-Inhibitory Separation. arXiv:2509.15460, 2025. [arXiv] |
| [A1] | Jin Hyun Park. Representations Learnt by SGD and Adaptive Learning Rules: Conditions that Vary Sparsity and Selectivity in Neural Networks. arXiv:2201.11653, 2022. [arXiv] |
Experience
| 2026 โ | Physical AI Ph.D. Fellow, Doosan Corporation, Seoul, Korea |
| 2022 โ | Research and Teaching Assistant, Texas A&M University, College Station, TX |
| 2023, 2025 | Machine Learning Engineer Intern, Tilda Corp., Seoul, Korea |
| 2024 | Machine Learning Engineer Intern, Samsung Life Insurance, Seoul, Korea |
| 2024 | Machine Learning Engineer Intern, LG Uplus (previously LG Telecom), Seoul, Korea |
| 2022 | Research Assistant, Kyungpook National University, Daegu, Korea |
| 2021 | Business Strategy Intern, Hyundai Motor Company, Seoul, Korea |
Education
| 2022 โ 2027 | Ph.D., Computer Science and Engineering, Texas A&M University (expected May 2027). Advisor: Yoonsuck Choe. |
| 2018 โ 2021 | B.Sc., Computer Science, First Class Honours, University of Bristol. Advisors: Rui Ponte Costa and Cian O'Donnell. |
Teaching and Service
| Reviewer / PC | AAAI 2024, 2025, 2026; AAAI-AIA 2026; AAAI-Demo 2026; IEEE TCDS 2025; NeurIPS NeurReps Workshop 2026. |
| Teaching assistant | CSCE 420/625 Artificial Intelligence, Texas A&M University. |
My Instagram account was hacked and I have not returned. If someone on Instagram with my name and face asks you for anything, it is not me.