Changhoon Kim

Bio

Changhoon Kim is an Assistant Professor at Soongsil University, specializing in robust and reliable AI. Previously, he was a Postdoctoral Scientist with the AWS Bedrock Science team, where he focused on enhancing the robustness of generative models. He completed his Ph.D. in Computer Engineering at Arizona State University under the advisement of Professor Yezhou Yang. His primary research focuses on the creation of secure machine learning systems. He has dedicated his efforts to developing user-attribution methods for generative models, a critical area of research in the age of AI-generated hyper-realistic content for tracing malicious usage, and machine UNlearning for removing private or harmful content from AI models. Kim's pioneering research has been recognized at prestigious conferences such as ICLR, ICML, ECCV, and CVPR. Additionally, he holds a U.S. patent for user-attribution in generative models. To further contribute to the community, he organizes tutorials and workshops at leading conferences to emphasize the importance of secure generative AI.

Research Keywords

- Computer Vision

- Generative Models

- Attribution of Synthesized Data

- Robust and Reliability for Generative Models 

News

Mar. 2025 Selected as Outstanding Reviewer for WACV'25. 

Mar. 2025 I started my role as an Assistant Professor.

Feb. 2025 I gave a talk to the AWS Guardrail Team.

Dec. 2024 Not All Adapters MAtter is accepted at NAACL.

Nov. 2024                TripletCLIP is accepted at NeurIPS.

Aug. 2024                R.A.C.E. received an ORAL presentation slot at ECCV.

Jul. 2024 As an organizer, our NeurIPS Workshop on Responsibly Building the Next Generation of Multimodal Foundation Models has been accepted.

Jul. 2024     R.A.C.E. is accepted at ECCV.

Jun. 2024       I have successfully defended my thesis.

Professional Experience

• Assistant Professor at Soongsil University

Seoul, ROK, 2025/03- 

• Post-doctoral Scientist at Amazon Web Service, Bedrock Team 

Arlington, VA, 2024/08 - 2025/02

• Ph.D. at Arizona State University

Tempe, AZ, 2018/08 - 2024/08

Invited Talks

• AWS Guardrail Team (Feb. 2025)

Topic: A Comprehensive Survey on Concept Erasure in Text-to-Image Diffusion Models

• AWS Customer Service Team (Dec. 2024)

Topic: Strengthening Image Generative AI

• University of Maryland, Baltimore County - PPR Seminar (Nov. 2024)

Topic: Strengthening Image Generative AI

• Seoul National University of Science and Technology (Nov. 2024)

Topic: Strengthening Image Generative AI

• ECCV Doctoral Consortium (Oct. 2024)

Topic: Robust Concept Erasure for T2I Models

• ASU Frontiers of GenAI (Sep. 2024)

Topic: Strengthening Image Generative AI: Integrating Fingerprinting and Revision Methods for Enhanced Safety and Control

• CVPR Doctoral Consortium (Jun. 2024)

Topic: Risk Management in Image Generative Models

• Samsung AI Center - Toronto (Apr. 2024)

Topic: Strengthening Image Generative AI

• AAAI Doctoral Consortium (Feb. 2024)

Topic: Risk Management in Image Generative Models through Model Fingerprinting

• Interview with AIhub (Jan. 2024)

Topic: Enhancing the reliability of image generative AI

• Mayo Clinic (July 2023)

Topic: Generative Model's Risk Management in Medical Domain

• Intel Labs (June 2023)

Topic: Text-to-Image Generative Model's Risk Management with User Attribution

• Korea University (July 2022)

Intelligent Architecture & Systems Research lab. 

Topic: Decentralized Attribution of Generative Models

Work Experience

• Research Scientist (Summer Intern)

KRAFTON, South Korea, 2022 Summer

• Computer Vision Engineer

GEOMania Co., South Korea, 2011-2014