Hyosoon Jang profile photo

Hyosoon Jang 장효순

Ph.D student

hyosoon.jang@kaist.ac.kr

About Me

I'm Hyosoon Jang, a Ph.D. student advised by Sungsoo Ahn at KAIST (2025.09–; previously at POSTECH, 2023.03–2025.09). My research focuses on drug discovery, with recent work on foundation models for molecular learning [P2,P3] and prior work on generative modeling approaches including Large Language Models (LLMs) [P1,C4], Generative Flow Networks (GFlowNets) [C2,C3], Diffusion Probabilistic Models (DPMs) [C1], and Variational Autoencoders (VAEs) [J2]. I am also interested in machine learning for medical domains [J1].

Recently, I have been participating in the Korean AI-specialized foundation model project (특파모), K-Fold, since October 2025. The project is supported by a large-scale computing environment with 256× NVIDIA B200 GPUs. In this project, I lead the development of one of the main components: protein representation learning methods for protein co-folding [P3].

Publications

indicates equal contribution.

[P3] Atom-level Protein Representation Learning Improves Protein Structure Prediction thumbnail

[P3] Atom-level Protein Representation Learning Improves Protein Structure Prediction

Hyosoon Jang, Taewon Kim, Hyunjin Seo, Seonghwan Seo, Hyeongwoo Kim, Wonho Zhung, Mingyeong Shin, Wooyoun Kim, Sungsoo Ahn

Preprint

[P2] A Systematic Evaluation of Co-folding Model Representations for Small-Molecule Learning thumbnail

[P2] A Systematic Evaluation of Co-folding Model Representations for Small-Molecule Learning

Hyosoon Jang, Hyunjin Seo, Honghui Kim, Seonghyun Park, Taewon Kim, Yunhui Jang, Sungsoo Ahn

Preprint

[C4] Self-Training Large Language Models with Confident Reasoning thumbnail

[C4] Self-Training Large Language Models with Confident Reasoning

Hyosoon Jang, Yunhui Jang, Sungjae Lee, Jungseul Ok, Sungsoo Ahn

EMNLP 2025 (findings)

[P1] Can LLMs Generate Diverse Molecules? Towards Alignment with Structural Diversity thumbnail

[P1] Can LLMs Generate Diverse Molecules? Towards Alignment with Structural Diversity

Hyosoon Jang, Yunhui Jang, Jaehyung Kim, Sungsoo Ahn

Preprint

[C3] Pessimistic Backward Policy for GFlowNets thumbnail

[C3] Pessimistic Backward Policy for GFlowNets

Hyosoon Jang, Yunhui Jang, Minsu Kim, Jinkyoo Park, Sungsoo Ahn

NeurIPS 2024

[C2] Learning Energy Decomposition for Partial Inference in GFlowNets thumbnail

[C2] Learning Energy Decomposition for Partial Inference in GFlowNets

Hyosoon Jang, Minsu Kim, Sungsoo Ahn

ICLR 2024 (oral presentation, 85/7262=1.16%)

[C1] Diffusion Probabilistic Models for Structured Node Classification thumbnail

[C1] Diffusion Probabilistic Models for Structured Node Classification

Hyosoon Jang, Seonghyun Park, Sangwoo Mo, Sungsoo Ahn

NeurIPS 2023

[J2] De novo Drug Design through Gradient-based Regularized Search in Information-theoretically Controlled Latent Space thumbnail

[J2] De novo Drug Design through Gradient-based Regularized Search in Information-theoretically Controlled Latent Space

Hyosoon Jang, Sangmin Seo, Sanghyun Park, Byung Ju Kim, Geon-Woo Choi, Jonghwan Choi, Chihyun Park

Journal of Computer-Aided Molecular Design 38 (1)

[J1] Machine learning algorithms using systemic inflammatory markers for predicting the oncological outcomes of colorectal cancer after surgery thumbnail

[J1] Machine learning algorithms using systemic inflammatory markers for predicting the oncological outcomes of colorectal cancer after surgery

Songsoo Yang, Hyosoon Jang, Inkyu Park, Sunhye Lee, Gaeul Oh, Chihyun Park*, Jeonghyun Kang*

Annals of Surgical Oncology 30 (13)

Invited Talks

UMICH, Stella group, GFlowNets, Jan 2025
ICLR, Oral session, Learning Energy Decompositions for Partial Inference in GFlowNets, May 2024
MILA, GFlowNet reading group, Learning Energy Decompositions for Partial Inference in GFlowNets, Apr 2024

Honors and Awards

Postechian Fellowship, 2024
Presidential Science Scholarship, 2024
National Excellence Scholarship (Natural Sciences and Engineering), 2021-2022
Merit-based Scholarship, 2019-2020