Siwon Kim
I am currently doing PhD course in Seoul National University, Korea. My research focuses on making AI models trustworthy. I deal with various types of data, including vision, natural language, and time-series signal. You can find my most recent CV, LinkedIn, and Google scholar here. Please contact me via tuslkk17@gmail.com :)
Research Interests
Responsible AI
Visually and textually explain deep learning models
Fair and safe image/text generation
Probing privacy leakage in large language models
Multi-modal Learning
Ground natural language to the physical world
News
2023-09-21 Our paper "ProPILE: Probing Privacy Leakage in Large Language Models" got accepted to NeurIPS 2023 as Spotlight!
2023-07-27 Selected as a Youlchon AI Star scholarship recipient!
2023-06-23 Our paper "De-stereotyping Text-to-image Models through Prompt Tuning" got accepted to ICML 2023 Workshop (DeployableGenerativeAI)!
2023-06-18 Started a return internship at Amazon, Seattle! I will stay at Seattle until the end of September.
2023-04-24 Two papers got accepted to ICML2023!
2023-02-28 Our paper "Grounding Counterfactual Explanation of Image Classifier to Textual Concept Space" got accepted to CVPR2023! (Work done at Amazon.)
2023-02-13 Started an internship at Parameter Lab! I will be interning until the beginning of June.
2022-07-08 Our paper "Grounding Visual Representations with Texts for Domain Generalization" got accepted to ECCV2022!
2022-06-20 Started an internship at Amazon, Seattle! I will be interning until the end of September.
Education
2018~Current Integrated Ph.D., Data Science and Artificial Intelligence Laboratory, Seoul National University
2014~2018 B.S., Electrical and Computer Engineering, Seoul National University
Working experiences
2023 Summer Applied Scientist Internship @ Amazon Alexa AI, Seattle
2023 Spring Research Internship @ ParameterLab, remote
2022 Summer Applied Scientist Internship @ Amazon Alexa AI, Seattle
Publications - Conference
2023 NeurIPS Siwon Kim, Sangdoo Yun, Hwaran Lee, Martin Gubri, Sungroh Yoon, and Seong Joon Oh, "ProPILE: Probing Privacy Leakage in Large Language Models"
2023 ICMLW Eunji Kim*, Siwon Kim*, Chaehun Shin, and Sungroh Yoon, "De-stereotyping Text-to-image models through Prompt Tuning" (*co-first)
2023 ICML Hyeongrok Han, Siwon Kim, Hyun-Soo Choi, and Sungroh Yoon, "On the Impact of Knowledge Distillation to Model Interpretability"
2023 ICML Eunji Kim, Dahuin Jung, Sangha Park, Siwon Kim, and Sungroh Yoon, "Probabilistic Concept Bottleneck model"
2023 CVPR Siwon Kim, Jinoh Oh, Sungjin Lee, Seunghak Yu, Jaeyoung Do, and Tara Taghavi, "Grounding counterfactual explanation of image classifiers to textual concept space"
2022 ECCV Seonwoo Min, Nokyung Park, Siwon Kim, Seunghyun Park, and Jinkyu Kim, "Grounding Visual Representations with Texts for Domain Generalization"
2022 CVPR Eunji Kim, Siwon Kim, Jungbeom Lee, Hyunwoo Kim, and Sungroh Yoon, "Bridging the gap between classification and localization for weakly supervised object localization"
2022 AAAI Siwon Kim, Kukjin Choi, Hyun-Soo Choi, Byunghan Lee, and Sungroh Yoon, "Towards a rigorous evaluation of time-series anomaly detection"
2021 CVPR Eunji Kim, Siwon Kim, Minji Seo, and Sungroh Yoon, "XProtoNet: Diagnosis in Chest Radiography with Global and Local Explanations" (Oral)
2020 EMNLP Siwon Kim, Jihun Yi, Eunji Kim, and Sungroh Yoon, "Interpretation of NLP models through input marginalization''
Publications - Journal
2022 TNNLS Hyun-Soo Choi, Dahuin Jung, Siwon Kim, and Sungroh Yoon, "Imbalanced Data Classification via Cooperative Interaction between Classifier and Generator''
2021 Access Seonwoo Min, Seunghyun Park, Siwon Kim, Hyun-Soo Choi, and Sungroh Yoon, "Pre-Training of Deep Bidirectional Protein Sequence Representations wit Structural Information''
2019 Access Siwon Kim*, Yonghoon Jeon*, Hyun-Soo Choi, Yoon Gi Chung, Sun Ah Choi, Hunmin Kim, Sungroh Yoon, Hee Hwang, and Ki Joong Kim, "Pediatric Sleep Stage Classification using Multi-domain Hybrid Neural Networks'' (*co-first)
Preprint
2020 ArXiv Jihun Yi, Eunji Kim, Siwon Kim, and Sungroh Yoon, "Information-theoretic visual explanation for black-box classifiers"
Project
2022~ Artificial Intelligence Reliability Research Center (CRC)
2019~ Explainable AI for Time-series Data w/ Hyundai Motor Company
2019~2020 AI Consortium, Hyundai AIRLab
Scholarship
2023 Youlchon AI Star
2022 Association for the Advancement of Artificial Intelligence (AAAI) scholarship
2014~2018 National Science & Technology Scholarship
Invited Talk
2023 "Introduction to Explainable AI" @ SeoulTech
2022 "Introduction to Explainable AI" @ Pusan University
2022 AI Retreat Poster Presentation @ Seoul National University
2021 "Interpretation of NLP models through input marginalization" @ Naver Tech. Talk
2021 "Advanced Data Scientist Course - XAI" @ Samsung Card