Identifies challenge sampling as a practical fault-attack surface in ML-DSA and HAETAE, and studies fault-based key recovery and countermeasures.
WonGeun Shin
Cryptographic Implementation Security
I am a Ph.D. candidate at Korea University, advised by Prof. HeeSeok Kim. My research focuses on physical attacks and countermeasures for cryptographic implementations, with a focus on post-quantum cryptography and secure neural-network implementations.
Selected Research
Develops first-order masked activation and ArgMax gadgets for side-channel resistant neural-network implementations.
Research Interests
Side-Channel Analysis
Leakage analysis and security evaluation of embedded cryptographic implementations.
Fault Attacks on PQC
Fault injection attacks and countermeasures for lattice-based digital signatures, including ML-DSA and HAETAE.
Masking & Secure Inference
Masking techniques for activation and ArgMax gadgets in side-channel resistant neural-network implementations.
News
Public Coefficient Matters is available as a preprint on IACR ePrint.
ARMOR has been published in IACR TCHES, vol. 2026, no. 2, pp. 106–131.
Working on fault injection attacks targeting ML-DSA and HAETAE.
Research Code
ARMOR
Explore the implementation accompanying our research on first-order masking for neural-network activation and ArgMax gadgets.
Contact
Email: shinryan9@korea.ac.kr
GitHub: github.com/persShins
Affiliation: Korea University