Welcome!

I am a fourth-year Ph.D. candidate in Computer Science at the University of Illinois at Urbana-Champaign advised by Prof. Julia Hockenmaier. My research primarily centers around:

• Improving our understanding of the inner workings of LLMs (e.g., mechanistic interpretability).
• Applying SAEs for practical purposes.
• Creating and analyzing novel deep learning model architectures (e.g., mixture-of-experts).

Education

• University of Illinois at Urbana-Champaign               Aug 2021 ~ Current
  • Ph.D. student in Computer Science
• Seoul National University                                            Mar 2013 ~ Feb 2021
  • B.S. in Electrical and Computer Enginering

Fellowships and Awards

• CS Ph.D. Fellowship (UIUC)
  • Sep 2023 - May 2024
  • Sep 2024 - May 2025
• Conference Presentation Awards for Graduate Students (UIUC)
  • FA24
• Excellent TA Award (UIUC)
  • FA23, FA24

Publications

1. Anonymous
   Under Review at COLM 2025
   Ikhyun Cho and Julia Hockenmaier
   
   
2. Anonymous
   Under Review at ACL 2025
   Ikhyun Cho, Gaeul Kwon, and Julia Hockenmaier
   
   
3. The Power of Bullet Lists: Reducing Mistakes in Large Language Models with a Simple Primer
   NAACL 2025 Findings
   Ikhyun Cho, Changyeon Park, and Julia Hockenmaier
   
   
4. Tutor-ICL: Guiding Large Language Models for Improved In-Context Learning Performance
   EMNLP 2024 Findings
   Ikhyun Cho, Gaeul Kwon, and Julia Hockenmaier
   
   
5. SIR-ABSC: Incorporating Syntax into RoBERTa-based Sentiment Analysis Models with a Special Aggregator Token
   EMNLP 2023 Findings
   Ikhyun Cho, Yoonhwa Jung, and Julia Hockenmaier
   
   
6. VisualSiteDiary: A Detector-Free Vision-Language Transformer Model for Captioning Photologs for Daily Construction Reporting and Image Retrievals
   Elsevier 2024: Automation in Construction
   Yoonhwa Jung, Ikhyun Cho, and Julia Hockenmaier
   
   
7. Pea-KD: Parameter-efficient and accurate Knowledge Distillation on BERT
   PLOS ONE 2022
   Ikhyun Cho and U Kang
   
   
8. SensiMix: Sensitivity-Aware 8-bit index & 1-bit value mixed precision quantization for BERT compression
   PLOS ONE 2022
   Tairen Piao, Ikhyun Cho, and U Kang

Unfortunate Publications 😭

1. Duplicate-and-Share: A Novel Approach to Efficient Vision Transformer Unlearning
   Ikhyun Cho, Changyeon Park, and Julia Hockenmaier
   
   
2. Prompting for Mixture-of-Experts: A Prompt-based Mixture-of-Experts framework for Stylized Image Captioning
   Ikhyun Cho, Yoonhwa Jung, and Julia Hockenmaier