About me

I am a third year PhD student at KAIST in Machine Learning and Artificial Intelligence (MLAI) Lab, under Prof. Sung Ju Hwang. I have been working on the intersections of AI and life science, bridging their gap by modeling biomolecules and their complex system via representation learning, generative models, and LLMs. [CV]

Email: cleverki [at] kaist [dot] ac [dot] kr

Publications

• Principia: Test-Time Hypotheses to Principle Construction for Scientific Discovery

  
  ​Dongki Kim*, Sangwoo Park*, Jinheon Baek, Soyeong Jeong, Sung Ju Hwang
  arXiv preprint, 2026

• AutoSciBench: Autonomous Benchmark Generation for Evaluating Scientific Agents

  [paper]
  ​Dongki Kim, Namkyeong Lee, Surag Nair, Carl Edwards, Xiner Li, Edward De Brouwer, Jenna Lynn Collier, Sung Ju Hwang, Gabriele Scalia, Ehsan Hajiramezanali
  arXiv preprint, 2026

• AdaPath: Query-Adaptive Path-Finding via Path-Bank for Multi-Hop Implicit Biomedical KGQA

  [paper]
  ​Jun Hyeong Kim, Dongki Kim, Yinhua Piao, Sung Ju Hwang
  Conference on Empirical Methods in Natural Language Processing (EMNLP), 2026

• ES-Merging: Biological MLLM Merging via Embedding Space Signals

  [paper]
  ​Wonbin Lee*, Dongki Kim*, Sung Ju Hwang
  Conference on Neural Information Processing Systems (NeurIPS), 2026

• Rethinking Reward Models for Multi-Domain Test-Time Scaling

  [paper]
  ​Dong Bok Lee*, Seanie Lee*, Sangwoo Park, Minki Kang, Jinheon Baek, Dongki Kim, Dominik Wagner, Jiongdao Jin, Heejun Lee, Tobias Bocklet, Jinyu Wang, Jingjing Fu, Sung Ju Hwang, Jiang Bian, Lei Song
  Transactions on Machine Learning Research (TMLR), 2026

• Multimodal Prompt Optimization: Why Not Leverage Multiple Modalities for MLLMs

  [paper]
  ​Yumin Choi*, Dongki Kim*, Jinheon Baek, Sung Ju Hwang
  International Conference on Learning Representations (ICLR), 2026

• Mol-LLaMA: Towards General Understanding of Molecules in Large Molecular Language Model

  [paper] [project page]
  ​Dongki Kim, Wonbin Lee, Sung Ju Hwang
  Conference on Neural Information Processing Systems (NeurIPS), 2025
  AI for Science at NeurIPS (AI4Science @ NeurIPS), 2025
  

• Graph Generation with Diffusion Mixture

  [paper]
  ​Jaehyeong Jo*, Dongki Kim*, Sung Ju Hwang
  International Conference on Machine Learning (ICML), 2024
  Machine Learning for Drug Discovery Workshop at ICLR (MLDD @ ICLR), 2023 (Spotlight)

• Protein Representation Learning by Capturing Protein Sequence-Structure-Function Relationship

  [paper]
  ​Eunki Ko*, Seul Lee*, Minseon Kim*, Dongki Kim, Sung Ju Hwang*
  ICLR Machine Learning for Genomics Explorations Workshop (MLGenX @ ICLR), 2024 (Spotlight)

• Antibody-SGM: Antigen-Specific Joint Design of Antibody Sequence and Structure using Diffusion Models

  [paper]
  ​ Xuezhi Xie, Jin Sub Lee, Dongki Kim, Jaehyeong Jo, Jisun Kim, Philip M. Kim
  Computation Biology Workshop at ICML (CompBio @ ICML), 2023

• Graph Self-supervised Learning with Accurate Discrepancy Learning

  [paper]
  ​Dongki Kim*, Jinheon Baek*, Sung Ju Hwang
  Conference on Neural Information Processing Systems (NeurIPS), 2022

• Edge Representation Learning with Hypergraphs

  [paper]
  ​Jaehyeong Jo*, Jinheon Baek*, Seul Lee*, Dongki Kim, Minki Kang, Sung Ju Hwang
  Conference on Neural Information Processing Systems (NeurIPS), 2021