Suho Kang

I am an Evaluation & Analysis Engineer at the Semiconductor R&D Center of Samsung Electronics. Concurrently, I serve as a part-time visiting researcher at the MIR Lab at Yonsei University, where I am supervised by Prof. Youngjae Yu.

My prior experience includes working as a part-time research assistant at the MLAI Lab under the guidance of of Prof. Kyungwoo Song, which significantly expanded my research expertise. I also earned my M.Eng. in Artificial Intelligence from Yonsei University as a part-time student while working full-time at Samsung Electronics.

My research interests lie in personalized and reliable AI, focusing on adapting LLMs and VLMs to open-world scenarios to enhance robustness and trustworthiness. Recently, my research has focused on enabling reliable human–LLM/VLM interaction and developing techniques for building small yet dependable models.

🔭 Publication

• RPLKG: Robust Prompt Learning with Knowledge Graph
  Yongtaek Lim∗, Yewon Kim∗, Suho Kang, Dokyung Yoon, Kyungwoo Song
  arXiv preprint 📄 Paper (*: Equal contribution)

• Benchmarking Foundation Models on Exceptional Cases: Dataset Creation and Validation
  Suho Kang, Jungyang Park, Joonseo Ha, SoMin Kim, JinHyeong Kim, Subeen Park, Kyungwoo Song
  EMNLP 2024 Workshop: GenBench 📄 Paper
  🏆 Recipient of the .Amazon Travel Scholarship Sponsored by

• Efficient Transfer Learning driven by Layer-wise Features Aggregation
  Chanwoo Kim, Jeyoon Yeom, Joowang Kim, Suho Kang, Kyungwoo Song
  NeurIPS 2024 Workshop: Adaptive Foundation Models 📄 Paper