About me

UPDATE

• As of Mar. 2025, I have returned to my Ph.D. program at KAIST after completing my alternative military service.

• During Oct. 2023 ~ Mar. 2025, I work as a research scientist at KRAFTON, as a continuation of the alternative of mandatory military service.

• During Mar. 2022 ~ Oct. 2023, I worked as a research scientist at AITRICS as the alternative for mandatory military service, working under the supervision of Prof. Sung Ju Hwang.

I am a Ph.d. student at Machine Learning and Artificial Intelligence (MLAI) lab in KAIST, under the supervision of Prof. Sung Ju Hwang.

My research interest includes:

• Efficient Deep Learning with Limited Data
  • Domain Adaptation and Generalization
• Natural Language Processing and Understanding
• Graph Representation Learning
• Text-to-Speech Synthesis

Education

• Korea Advanced Institute of Science and Technology (KAIST)
  Sep. 2021 -

  Ph.d. in Graduate School of Artificial Intelligence

• Korea Advanced Institute of Science and Technology (KAIST)
  Mar. 2020 - Aug. 2021

  M.S. in Graduate School of Artificial Intelligence

• Korea Advanced Institute of Science and Technology (KAIST)
  Mar. 2015 - Feb. 2020

  B.S. in Electrical Engineering and Computer Science

Preprints

• [P2] Distilling LLM Agent into Small Models with Retrieval and Code Tools
  [paper]
  Minki Kang, Jongwon Jeong, Seanie Lee, Jaewoong Cho, Sung Ju Hwang \

• [P1] T1: Tool-integrated Self-verification for Test-time Compute Scaling in Small Language Models
  [paper]
  Minki Kang*, Jongwon Jeong*, Jaewoong Cho \

Conference Publications

• [C18] SafeRoute: Adaptive Model Selection for Efficient and Accurate Safety Guardrails in Large Language Models
  [paper]
  Seanie Lee*, Dong Bok Lee*, Dominik Wagner, Minki Kang, Haebin Seong, Tobias Bocklet, Juho Lee, Sung Ju Hwang
  ACL 2025 Findings

• [C17] HarmAug: Effective Data Augmentation for Knowledge Distillation of Safety Guard Models
  [paper]
  Seanie Lee*, Haebin Seong*, Dong Bok Lee, Minki Kang, Xiaoyin Chen, Dominik Wagner, Yoshua Bengio, Juho Lee, Sung Ju Hwang
  ICLR 2025

• [C16] Stable-TTS: Stable Speaker-Adaptive Text-to-Speech Synthesis via Prosody Prompting
  [paper]
  Wooseok Han*, Minki Kang*, Changhun Kim, Eunho Yang
  ICASSP 2025

• [C15] Face-StyleSpeech: Enhancing Zero-shot Speech Synthesis from Face Images with Improved Face-to-Speech Mapping
  [paper]
  Minki Kang*, Wooseok Han*, Eunho Yang
  ICASSP 2025

• [C14] Face-StyleSpeech: Enhancing Zero-shot Speech Synthesis from Face Images with Improved Face-to-Speech Mapping
  [paper]
  Minki Kang, Wooseok Han, Eunho Yang
  ICASSP 2025

• [C13] Latent Paraphrasing: Perturbation on Layers Improves Knowledge Injection in Language Models
  [paper]
  Minki Kang, Sung Ju Hwang, Gibbeum Lee, Jaewoong Cho
  NeurIPS 2024

• [C12] Knowledge-Augmented Language Model Verification
  [paper]
  Jinheon Baek, Soyeong Jeong, Minki Kang, Jong C. Park, Sung Ju Hwang
  EMNLP 2023

• [C11] Knowledge-Augmented Reasoning Distillation for Small Language Models in Knowledge-Intensive Tasks
  [paper]
  Minki Kang, Seanie Lee, Jinheon Baek, Kenji Kawaguchi, Sung Ju Hwang
  NeurIPS 2023

• [C10] ZET-Speech: Zero-shot adaptive Emotion-controllable Text-to-Speech Synthesis with Diffusion and Style-based Models
  [paper]
  Minki Kang*, Wooseok Han*, Sung Ju Hwang, Eunho Yang
  Interspeech 2023

• [C9] Grad-StyleSpeech: Any-speaker Adaptive Text-To-Speech Synthesis with Diffusion Models
  [paper]
  Minki Kang*, Dongchan Min*, Sung Ju Hwang
  ICASSP 2023

• [C8] Sparse Token Transformer with Attention Back Tracking
  [paper]
  Heejun Lee, Minki Kang, Youngwan Lee, Sung Ju Hwang
  ICLR 2023

• [C7] Self-Distillation for Further Pre-training of Transformers
  [paper]
  Seanie Lee, Minki Kang, Juho Lee, Sung Ju Hwang, Kenji Kawaguchi
  ICLR 2023

• [C6] KALA: Knowledge-Augmented Language Model Adaptation
  [paper]
  Minki Kang*, Jinheon Baek*, Sung Ju Hwang
  NAACL 2022

• [C5] Edge Representation Learning with Hypergraphs
  [paper]
  Jaehyeong Jo*, Jinheon Baek*, Seul Lee*, Dongki Kim, Minki Kang, Sung Ju Hwang
  NeurIPS 2021

• [C4] Learning to Perturb Word Embeddings for Out-of-distribution QA
  [paper]
  Seanie Lee*, Minki Kang*, Juho Lee, Sung Ju Hwang
  ACL 2021

• [C3] Accurate Learning of Graph Representations with Graph Multiset Pooling
  [paper]
  Jinheon Baek*, Minki Kang*, Sung Ju Hwang
  ICLR 2021

• [C2] Neural Mask Generator: Learning to Generate Adaptive Word Maskings for Language Model Adaptation
  [paper]
  Minki Kang*, Moonsu Han*, Sung Ju Hwang
  EMNLP 2020

• [C1] Episodic Memory Reader: Learning What to Remember for Question Answering from Streaming Data
  [paper]
  Moonsu Han*, Minki Kang*, Hyunwoo Jung, Sung Ju Hwang
  ACL 2019

Workshop Publications

• [W1] Knowledge Graph-Augmented Language Models for Knowledge-Grounded Dialogue Generation
  [workshop] [paper]
  Minki Kang*, Jin Myung Kwak*, Jinheon Baek*, Sung Ju Hwang
  Knowledge Retrieval and Language Models (KRLM) Workshop @ ICML 2022

(*: equal contribution)

Experiences

• KRAFTON
  Oct. 2023 - Mar. 2025

  Natural Language DL Team @ AI Research Center

• AITRICS
  Mar. 2022 - Oct. 2023

  Researcher @ Virtual Human Team

• Technial University of Munich
  Apr. 2019 - Aug. 2019

  Exchange Student @ Electical and Computer Engineering Department

• Kakao
  Jan. 2018 - Feb. 2018

  Internship @ Context Department