About

I am an AI Researcher at Boeing and an incoming CS masters student at Logo. Previously, I received B.S. degree at Yonsei University majoring in Electrical and Electronic Engineering with a minor in Astronomy. My studies will be funded by Logo.

Currently, I am highly interested in:

  • Representation Learning
  • AI for Science
  • Causal AI

Alongside, I am keen to put effort into :

  • Compositional Understanding
  • Mechanistic Interpretability

CV Google Scholar hazelnam2000 at gmail dot com

News

May. 2025 I am honored to be selected as a Logo graduate student.
Mar. 2025 Our multimodal reasoning benchmark VAGUE 2.0 is released!
Oct. 2024 1 paper is accepted at AdvML-Frontiers @ NeurIPS 2024
Jun. 2024 I started research collaboration @ Causality Lab (SNU).
Jan. 2024 I joined Logo as an AI Researcher.
Dec. 2023 1 paper is accepted at ICASSP 2024
Oct. 2023 1 paper is accepted at CRL @ NeurIPS 2023
Oct. 2023 Selected as a NeurIPS 2023 Volunteer.
Sep. 2023 I started Industry-Academic Cooperation @ LinqAlpha.
Sep. 2023 1 paper is accepted in 4th Korea Artificial Intelligence Conference.
Oct. 2022 I starteed my internship at AITRICS (~Feb '23).
Jun. 2022 I starteed my internship at VRL @ UCSB. (~Sep '22)

Publications

* indicates equal contribution.


Causal Representation Learning with Partially Observable Sources
Kwonho Kim, Inwoo Hwang, Heejeong Nam, Sanghack Lee
Under Review

When AI Co-Scientists Fail: SPOT-a Benchmark for Automated Verification of Scientific Research
Guijin Son, Jiwoo Hong, Honglu Fan, Heejeong Nam, Hyunwoo Ko, Seungwon Lim, Jinyeop Song, Jinha Choi, Gonçalo Paulo, Youngjae Yu, Stella Biderman
Preprint

VAGUE: Visual Contexts Clarify Ambiguous Expressions
Heejeong Nam*, Jinwoo Ahn*, Keummin Ka, Jiwan Chung, Youngjae Yu
Preprint

An Adversarial Learning Approach to Irregular Time-Series Forecasting
Heejeong Nam, Jihyun Kim, Jimin Yeom
NeurIPS 2024 Workshop on AdvML-Frontiers

Compact and De-biased Negative Instance Embedding for Multi-Instance Learning on Whole-Slide Pathological Images
Joohyeong Lee, Heejeong Nam, Kwanhyeong Lee, Sangchul Hahn
International Conference on Acoustics, Speech and Signal Processing (ICASSP). IEEE, 2024

SCADI: Self-supervised Causal Disentanglement in Latent Variable Models
Heejeong Nam
NeurIPS 2023 Workshop on Causal Representation Learning

Enhanced Open Set Recognition via Disentangled Representation Learning
Heejeong Nam
2023 Korea Artificial Intelligence Conference pp.208-210

Past Experiences

Miscellaneous

I love (almost all) outdoor activities. I enjoy working out, figure skating, snowboarding, swimming, and bouldering. I'm always excited to try new sports. I also like taking pictures of nature during the travel.