Namhyoung Kim

Head of Research, Risk X tech Lab
Ph.D. in Industrial Engineering, Hanyang University

ai.namhyoung@gmail.com GitHub Google Scholar LinkedIn CV (PDF)

I am Head of Research at Risk X tech Lab, where I lead the research team, and received my Ph.D. in Industrial Engineering from Hanyang University in 2026, advised by Prof. Jae Wook Song at the Financial Innovation & Decision Analytics (FINX) Lab.

My research studies representation learning for low signal-to-noise data, with financial markets as the main testbed. I work on discrete representation learning, centered on vector-quantized latent factor models; on structure-informed learning, which builds known financial dynamics into neural models; and on interpretable and group-robust models.

My recent work on vector-quantized discrete latent factors appeared at IJCAI-ECAI 2026, and a co-authored paper on probabilistic volatility forecasting under distribution shift has been accepted to ICAIF 2026. I am always open to research discussions, so feel free to reach out.

News

2026-10 Our paper Text as Structure, Not Signal: Text-Aligned Groups for Distributionally Robust Stock Ranking has been accepted to IAFM ’26, the ACM ICAIF 2026 Workshop on Interpretability and Alignment of Financial Models.
2026-10 Our paper HAR-CEV: Probabilistic Volatility Forecasting with Adaptive Mean Reversion under Distribution Shift has been accepted to ICAIF ’26.
2026-08 Presented our IJCAI-ECAI 2026 paper on vector-quantized discrete latent factors in Bremen, Germany.
2026-07 Presented the same work at the Asian Quantitative Finance Conference (AQFC 2026) in Pohang.
2026-05 Our paper Vector-Quantized Discrete Latent Factors Meet Financial Priors: Dynamic Cross-Sectional Stock Ranking Prediction for Portfolio Construction has been accepted to IJCAI-ECAI 2026.
2026-03 Appointed Head of Research at Risk X tech Lab.
2026-02 Successfully defended my Ph.D. dissertation, Representation Learning in Financial Markets for Portfolio Management, at Hanyang University.
2025-11 Our paper Deep Mean-Reversion: A Physics-Informed Contrastive Approach to Pairs Trading was accepted to ICAIF ’25 as an oral presentation (acceptance rate 15.5%).
2025-07 Joined Risk X tech Lab as a Quantitative AI Researcher.
2025-04 FactorVQVAE has been published in Knowledge-Based Systems.
2024-11 Received the Outstanding Paper Award from Korean Management Science Review (KORMS, 2024).

Selected Publications

IJCAI '26

Vector-Quantized Discrete Latent Factors Meet Financial Priors: Dynamic Cross-Sectional Stock Ranking Prediction for Portfolio Construction

Namhyoung Kim, Jae Wook Song*

Proceedings of the 35th International Joint Conference on Artificial Intelligence (IJCAI-ECAI 2026), 2026 · pp. 4421–4429 · Acceptance rate 13.6% · CORE A*

Proceedings DOI arXiv

@inproceedings{kim2026vector,
  title     = {Vector-Quantized Discrete Latent Factors Meet Financial Priors: Dynamic Cross-Sectional Stock Ranking Prediction for Portfolio Construction},
  author    = {Kim, Namhyoung and Song, Jae Wook},
  booktitle = {Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence, {IJCAI-26}},
  pages     = {4421--4429},
  year      = {2026},
  publisher = {International Joint Conferences on Artificial Intelligence Organization},
  doi       = {10.24963/ijcai.2026/492}
}

ICAIF '26

HAR-CEV: Probabilistic Volatility Forecasting with Adaptive Mean Reversion under Distribution Shift

Yosep Na, Namhyoung Kim, Joon Ho Jun, Jae Wook Song*

Proceedings of the 7th ACM International Conference on AI in Finance (ICAIF '26), 2026 · Accepted, to appear

@inproceedings{na2026harcev,
  title     = {{HAR-CEV}: Probabilistic Volatility Forecasting with Adaptive Mean Reversion under Distribution Shift},
  author    = {Na, Yosep and Kim, Namhyoung and Jun, Joon Ho and Song, Jae Wook},
  booktitle = {Proceedings of the 7th ACM International Conference on AI in Finance},
  series    = {ICAIF '26},
  year      = {2026},
  publisher = {ACM},
  note      = {To appear}
}

ICAIF '25

Deep Mean-Reversion: A Physics-Informed Contrastive Approach to Pairs Trading

Namhyoung Kim, Yosep Na, Jae Wook Song*

Proceedings of the 6th ACM International Conference on AI in Finance (ICAIF '25), 2025 · pp. 405–412 · Oral · Acceptance rate 15.5%

ACM DL DOI

@inproceedings{kim2025deep,
  title     = {Deep Mean-Reversion: A Physics-Informed Contrastive Approach to Pairs Trading},
  author    = {Kim, Namhyoung and Na, Yosep and Song, Jae Wook},
  booktitle = {Proceedings of the 6th ACM International Conference on AI in Finance},
  series    = {ICAIF '25},
  pages     = {405--412},
  year      = {2025},
  publisher = {ACM},
  doi       = {10.1145/3768292.3770406}
}

KBS '25

FactorVQVAE: Discrete Latent Factor Model via Vector Quantized Variational Autoencoder

Namhyoung Kim, Seung Eun Ock, Jae Wook Song*

Knowledge-Based Systems, 2025 · Vol. 318, Article 113460 · IF 7.6 · JCR Q1 (Top 6%, AI / Computer Science)

DOI

@article{kim2025factorvqvae,
  title     = {{FactorVQVAE}: Discrete Latent Factor Model via Vector Quantized Variational Autoencoder},
  author    = {Kim, Namhyoung and Ock, Seung Eun and Song, Jae Wook},
  journal   = {Knowledge-Based Systems},
  volume    = {318},
  pages     = {113460},
  year      = {2025},
  publisher = {Elsevier},
  doi       = {10.1016/j.knosys.2025.113460}
}

All publications