Keynotes

Title: Secure Programming with Large Language Models: Are We There Yet?

Speaker: Hyoungshick Kim (Sungkyunkwan University)

Date: October 20, 2025
Abstract: AI coding assistants, such as GitHub Copilot, have quickly become an integral part of developers’ everyday workflows, but how safe is the code they generate? In this talk, I’ll dive into what makes AI-generated code insecure, drawing from recent case studies and experiments. I’ll walk through efforts to improve security using fine-tuning, prompt engineering, and retrieval-augmented techniques, as well as our own approach to automated patch generation using vulnerability logs. This talk highlights both the potential and the current gaps in making large language models truly helpful for secure programming. The goal is to move beyond fixing bugs to actually teaching models why code is secure.

Biography: Hyoungshick Kim is a professor in the Department of Computer Science and Engineering, Sungkyunkwan University. He received a BS degree from the Department of Information Engineering at Sungkyunkwan University, an MS from the Department of Computer Science at KAIST, and a Ph.D. from the Computer Laboratory at the University of Cambridge in 1999, 2001, and 2011, respectively. After completing his Ph.D., he worked as a post-doctoral fellow in the Department of Electrical and Computer Engineering at the University of British Columbia. He previously worked as a senior engineer at Samsung Electronics from 2004 to 2008. He also worked as a distinguished visiting researcher at CSIRO Data61 from 2019 to 2020. His current research interests include usable security, software security, and data-driven security. He enjoys finding security issues in new systems, particularly recent AI systems and applications, and has been deeply engaged in identifying real security problems in these areas. His work aims to uncover and address the practical security challenges posed by emerging technologies.

Title: Sokół: Polishing Up Falcon

Speaker: Mehdi Tibouchi (NTT Social Informatics Laboratories & Kyoto University)

Date: October 21, 2025
Abstract: Falcon is a post-quantum signature scheme based on lattices that has been selected by NIST for upcoming standardization. It is compact, efficient and a really attractive design overall, but presents a number of pitfalls in terms of implementation. In this talk, we will present some of these issues and how they can affect the security of real-world deployments. We will then suggest some little touch-ups to the scheme that largely mitigate those issues, and could usefully be considered in the final standard.

Biography: After obtaining his Ph.D. in computer science from Univ. Paris VII and Univ. Luxembourg in 2011, Mehdi Tibouchi joined NTT, Japan as a researcher in cryptography. He is now a distinguished researcher at the NTT Social Informatics Laboratories, and a guest associate professor at Kyoto University. His research interests cover various mathematical aspects of public-key cryptography and cryptanalysis, particularly related to elliptic curves and Euclidean lattices.