Seung Ho Na

When LLMs Go Online: The Emerging Threat of Web-Enabled LLMs

Hanna Kim, Minkyoo Song, Seung Ho Na, Seungwon Shin, Kimin Lee

PDF Bib To appear.

Obliviate: Neutralizing Task-agnostic Backdoors within the Parameter-efficient Fine-tuning Paradigm

Jaehan Kim, Minkyoo Song, Seung Ho Na, Seungwon Shin

PDF Bib To appear.

CENSor: Detecting Illicit Bitcoin Operation via GCN-based Hyperedge Classification

Suyeol Lee, Jaehan Kim, Minjae Seo, Seung Ho Na, Seungwon Shin, Jinwoo Kim

PDF Bib To appear.

Evolving Bots: The New Generation of Comment Bots and their Underlying Scam Campaigns in YouTube

Seung Ho Na , Sumin Cho, Seungwon Shin

PDF Bib @inproceedings{na2023evolving,
   title={Evolving Bots: The New Generation of Comment Bots and their Underlying Scam Campaigns in YouTube},
   author={Na, Seung Ho and Cho, Sumin and Shin, Seungwon},
   booktitle={Proceedings of the 2023 ACM on Internet Measurement Conference},
   pages={297–312},
   year={2023}
}

Witnessing Erosion of Membership Inference Defenses: Understanding Effects of Data Drift in Membership Privacy

Seung Ho Na , Kwanwoo Kim, Seungwon Shin

PDF Bib @inproceedings{na2023witnessing,
   title={Witnessing Erosion of Membership Inference Defenses: Understanding Effects of Data Drift in Membership Privacy},
   author={Na, Seung Ho and Kim, Kwanwoo and Shin, Seungwon},
   booktitle={Proceedings of the 26th International Symposium on Research in Attacks, Intrusions and Defenses},
   pages={250–263},
   year={2023}
}

Closing the Loophole: Rethinking Reconstruction Attacks in Federated Learning from a Privacy Standpoint

Seung Ho Na , Hyeong Gwon Hong, Junmo Kim, Seungwon Shin

PDF Bib @inproceedings{na2022closing,
   title={Closing the Loophole: Rethinking Reconstruction Attacks in Federated Learning from a Privacy Standpoint},
   author={Na, Seung Ho and Hong, Hyeong Gwon and Kim, Junmo and Shin, Seungwon},
   booktitle={Annual Computer Security Applications Conference},
   pages={332–345},
   year={2022}
}

Reconfigurable Regular Expression Matching Architecture for Real-time Pattern Update and Payload Inspection

Jaehyun Nam, Seung Ho Na, Seungwon Shin, Taejune Park

PDF Bib @article{nam2022reconfigurable,
   title={Reconfigurable regular expression matching architecture for real-time pattern update and payload inspection},
   author={Nam, Jaehyun and Na, Seung Ho and Shin, Seungwon and Park, Taejune},
   journal={Journal of Network and Computer Applications},
   volume={208},
   pages={103507},
   year={2022},
   publisher={Elsevier}
}

Meta-Path-based Fake News Detection Leveraging Multi-level Social Context Information

Jian Cui, Kwanwoo Kim, Seung Ho Na, Seungwon Shin

PDF Bib @inproceedings{cui2022meta,
   title={Meta-Path-based Fake News Detection Leveraging Multi-level Social Context Information},
   author={Cui, Jian and Kim, Kwanwoo and Na, Seung Ho and Shin, Seungwon},
   booktitle={Proceedings of the 31st ACM International Conference on Information \& Knowledge Management},
   pages={325–334},
   year={2022}
}

Reinhardt: Real-time Reconfigurable Hardware Architecture for Regular Expression Matching in DPI

Taejune park, Jaehyun Nam, Seung Ho Na, Jaewoong Chung, Seungwon Shin

PDF Bib @inproceedings{park2021reinhardt,
   title={Reinhardt: Real-time Reconfigurable Hardware Architecture for Regular Expression Matching in DPI},
   author={Park, Taejune and Nam, Jaehyun and Na, Seung Ho and Chung, Jaewoong and Shin, Seungwon},
   booktitle={Annual Computer Security Applications Conference},
   pages={620–633},
   year={2021}
}

[Poster] Knowledge Seeking on The Shadow Brokers

Seung Ho Na*, Kwanwoo Kim*, Seungwon Shin

*:Equally Contributed

PDF Bib @inproceedings{na2018knowledge,
   title={Knowledge Seeking on The Shadow Brokers},
   author={Na, Seung Ho and Kim, Kwanwoo and Shin, Seungwon},
   booktitle={Proceedings of the 2018 ACM SIGSAC Conference on Computer and Communications Security},
   pages={2249–2251},
   year={2018}
}

Worked as an intern in the research team. Was assigned a project on using deep learning for anomaly detection of faulty factory components, where I developed a hierarchical classification system that helped decrease false negative samples.

As an engineering student intern at a venture capital company, I had more familiarity in technology industries and conducted market research on IoT and renewable energy.

Under Professor Seungwon Shin, I was able to utilize my AI knowledge and apply it to CTI (cyber threat information) research. My Master’s thesis is research on a knowledge-powered cross-domain investigation of cyber-extortion operators. This work presents an analysis of Clop, a ransomware campaign, using a devised investigation framework that utilizes information from multiple domains.

During my time at KAIST I dedicated much of my time in exploring what I wanted to do in life. After a broad selection of courses from various fields, several undergraduate research programs, and extracurricular activities, I found that learning and discovering new knowledge was what I enjoyed and decided to continue this in graduate school.