Publications [Google Scholar]
Conf Journal Workshop Preprint
Preprints
Asking Forever: Universal Activations Behind Turn Amplification in Conversational LLMs
Zachary Coalson, Bo Fang, and Sanghyun Hong
arXiv Preprint. 2026. Preprint
Discovering Universal Activation Directions for PII Leakage in Language Models
Leo Marchyok, Zachary Coalson, Sungho Keum, Sooel Son, and Sanghyun Hong
arXiv Preprint. 2026. Preprint
Fail-Closed Alignment for Large Language Models
Zachary Coalson, Beth Sohler, Aiden Gabriel, and Sanghyun Hong
arXiv Preprint. 2026. Preprint
Data Therapist: Eliciting Domain Knowledge from Subject Matter Experts Using Large Language Models
Sungbok Shin, Hyeon Jeon, Sanghyun Hong, and Niklas Elmqvist
arXiv Preprint. 2025. Preprint
Modeling Neural Networks with Privacy Using Neural Stochastic Differential Equations
Sanghyun Hong, Fan Wu, Anthony Gruber, and Kookjin Lee
arXiv Preprint. 2025. Preprint
PrisonBreak: Jailbreaking Large Language Models with Fewer Than Twenty-Five Targeted Bit-flips
Zachary Coalson, Jeonghyun Woo, Shiyang Chen, Yu Sun, Lishan Yang, Prashant Nair, Bo Fang, and Sanghyun Hong
arXiv Preprint. 2024. Preprint
Certified Robustness to Clean-label Poisoning Using Diffusion Denoising
Sanghyun Hong, Nicholas Carlini, and Alexey Kurakin
arXiv Preprint. 2024. Preprint
On the Effectiveness of Regularization Against Membership Inference Attacks
Yiǧitcan Kaya, Sanghyun Hong, and Tudor Dumitraş
arXiv Preprint. 2020. Preprint
On the Effectiveness of Mitigating Data Poisoning Attacks with Gradient Shaping
Sanghyun Hong, Varun Chandrasekaran, Yiǧitcan Kaya, Tudor Dumitraș, and Nicolas Papernot
arXiv Preprint. 2020. Preprint
2026
Membership Inference in Fine-tuned Diffusion Language Models via Token-level Memorization Asymmetry
Shengfang Zhai, Leo Marchyok, Yuling Shi, Huanran Chen, Yinpeng Dong, Jiaheng Zhang, and Sanghyun Hong
Findings of the Conference on Empirical Methods in Natural Language Processing (EMNLP). 2026. Conf
Do Reasoning Representations Help Humans Evaluate LLM Outputs?
Jaewoo Lim, Sungbok Shin, and Sanghyun Hong
Findings of the Conference on Empirical Methods in Natural Language Processing (EMNLP). 2026. Conf
On the Resilience of Text-to-Video Diffusion Models to Hardware Faults
Zachary Coalson, A M Aahad, Stella Doehring, Zane Ma, and Sanghyun Hong
ICML Workshop on From Frames to Stories (F2S). 2026. Workshop
Adversarial Robustness of Implicit Neural Representation Based Classifiers
Jayoung Kim, Kookjin Lee, Noseong Park, and Sanghyun Hong
International Conference on Machine Learning (ICML). 2026. Conf
When Can You Poison Rewards? A Tight Characterization of Reward Poisoning in Linear MDPs
Jose Efraim Aguilar Escamilla, Haoyang Hong, Jiawei Li, Haoyu Zhao, Xuezhou Zhang, Sanghyun Hong, and Huazheng Wang
International Conference on Machine Learning (ICML). 2026. Conf
AgentBreaker: A Framework for Evaluating Context-Aware Indirect Prompt Injection Risks in Modern Web Agents
Yongbi Son, Changoo Lee, Dongwon Shin, Byoungyoung Lee, Sanghyun Hong, and Sooel Son
ACM SIGSOFT International Symposium on Software Testing and Analysis (ISSTA). 2026. Conf
Hessian-aware Training for Enhancing DNNs Resilience to Parameter Corruptions
Tahmid Hasan Prato, Seijoon Kim, Lizhong Chen, and Sanghyun Hong
Transactions on Machine Learning Research (TMLR). 2026. Journal
Site Isolation is Dead: How Site Isolation is Broken in Agentic Browsers and Extensions
Suyoung Lee, Seongho Keum, Changoo Lee, Dongwon Shin, Sanghyun Hong, Byoungyoung Lee, and Sooel Son
IEEE Symposium on Security and Privacy (IEEE S&P). 2026. Conf
Computational and Experimental Pathways to Next-Generation Ultrawide-Band-Gap Oxide Semiconductors
Sieun Chae, Jongin Kim, Joshua R. Anderson, Sanghyun Hong, Yaser Mike Banad, and Hanjong Paik
Nano Convergence. 2026. Journal
2025
Hard Work Does Not Always Pay Off: Poisoning Attacks on Neural Architecture Search
Zachary Coalson, Huazheng Wang, Qingyun Wu, and Sanghyun Hong
Transactions on Machine Learning Research (TMLR). 2025. Journal
Inference-Time Noise Addition for Improving Adversarial Robustness of Audio Deepfake Detection System
Inho Kim, Phuc Thien Doan, Sanghyun Hong, and Souhwan Jung
IEEE Access. 2025. Journal
When Does Wasm Malware Detection Fail? A Systematic Analysis of Their Robustness to Evasion
Taeyoung Kim, Sanghak Oh, Kiho Lee, Weihang Wang, Yonghwi Kwon, Sanghyun Hong, and Hyoungshick Kim
IEEE/ACM Automated Software Engineering Conference (ASE). 2025. Conf
Mind the Gap: Time-of-Check to Time-of-Use Vulnerabilities in LLM-Enabled Agents
Derek Lilienthal and Sanghyun Hong
NeurIPS Workshop on Machine Learning for Systems (MLforSystems). 2025. Workshop
IF-Guide: Influence Function-Guided Detoxification of LLMs
Zachary Coalson, Juhan Bae, Nicholas Carlini, and Sanghyun Hong
Advances in Neural Information Processing Systems (NeurIPS). 2025. Conf
Demystifying the Resilience of Large Language Model Inference: An End-to-End Perspective
Yu Sun, Zachary Coalson, Shiyang Chen, Hang Liu, Sanghyun Hong, Zhao Zhang, Bo Fang, and Lishan Yang
ACM/IEEE Supercomputing Conference (SC). 2025. Conf
Harnessing Input-adaptive Inference for Efficient VLN
Dongwoo Kang, Akhil Perincherry, Zachary Coalson, Aiden Gabriel, Stefan Lee, and Sanghyun Hong
International Conference on Computer Vision (ICCV). 2025. Conf
Evaluating Robustness of Reference-based Phishing Detectors
Eunjin Roh*, Sungwoo Jeon*, Sooel Son, and Sanghyun Hong (*equal contribution)
ACM ASIA Conference on Computer and Communication Security (AsiaCCS). 2025. Conf
Enhancing Audio Deepfake Detection by Improving Representation Similarity of Bonafide Speech
Seung-bin Kim, Hyun-seo Shin, Jungwoo Heo, Chan-yeong Lim, Kyo-Won Koo, Jisoo Son, Sanghyun Hong, Souhwan Jung, and Ha-Jin Yu
Interspeech. 2025. Conf
Private Investigator: Extracting Personally Identifiable Information from Large Language Models Using Optimized Prompts
Seongho Keum, Dongwon Shin, Leo Marchyok, Sanghyun Hong, and Sooel Son
USENIX Security Symposium (USENIX Security). 2025. Conf
Evaluating Memorization in Parameter-Efficient Fine-tuning
Sanghyun Hong, Nicholas Carlini, and Alexey Kurakin
ICML Workshop on the Impact of Memorization on Trustworthy Foundation Models (MemFM). 2025. [Oral] Workshop
MADCAT: Combating Malware Detection Under Concept Drift with Test-Time Adaptation
Eunjin Roh, Yigitcan Kaya, Christopher Kruegel, Giovanni Vigna, and Sanghyun Hong
ICML Workshop on Test-Time Adaptation (PUT). 2025. Workshop
Visualizationary: Automating Design Feedback for Visualization Designers using LLMs
Sungbok Shin, Sanghyun Hong, and Niklas Elmqvist
IEEE Transactions on Visualization and Computer Graphics (IEEE VIS). 2025. Conf
SoK: Watermarking for AI-Generated Content
Xuandong Zhao, Sam Gunn, Miranda Christ, Jaiden Fairoze, Andres Fabrega, Nicholas Carlini, Milad Nasr, Sanghyun Hong, Florian Tramer, Sanjam Garg, Somesh Jha, Lei Li, Yu-Xiang Wang, and Dawn Song
IEEE Symposium on Security and Privacy (IEEE S&P). 2025. Conf
2024
A Stability Analysis of Neural Networks and Its Application to Tsunami Early Warning
Donsub Rim, Sanah Suri, Sanghyun Hong, Kookjin Lee, and Randall J LeVeque
Journal of Geophysical Research: Machine Learning and Computation. 2024. Journal
You Never Know: Quantization Induces Inconsistent Biases in Vision-Language Foundation Models
Eric Slyman, Anirudh Kanneganti, Sanghyun Hong, and Stefan Lee
NeurIPS Workshop on Responsibly Building the Next Generation of Multimodal Foundational Models (RBFM). 2024. Workshop
Privacy Backdoors: Enhancing Membership Inference through Poisoning Pre-trained Models
Yuxin Wen, Leo Marchyok, Sanghyun Hong, Jonas Geiping, Tom Goldstein, and Nicholas Carlini
Advances in Neural Information Processing Systems (NeurIPS). 2024. Conf
You Only Perturb Once: Bypassing (Robust) Ad-Blockers Using Universal Adversarial Perturbations
Dongwon Shin, Suyoung Lee, Sanghyun Hong, and Sooel Son
Annual Computer Security Applications Conference (ACSAC). 2024. Conf
Identifying Contemporaneous and Lagged Dependence Structures by Promoting Sparsity in Continuous-time Neural Networks
Fan Wu, Woojin Cho, David Korotky, Sanghyun Hong, Donsub Rim, Noseong Park, and Kookjin Lee
ACM International Conference on Information and Knowledge Management (CIKM). 2024. Conf
LeaPformer: Enabling Linear Transformers for Autoregressive and Simultaneous Tasks via Learned Proportions
Victor Agostinelli III, Sanghyun Hong, and Lizhong Chen
International Conference on Machine Learning (ICML). 2024. Conf
Parameterized Physics-informed Neural Networks for Parameterized PDEs
Woojin Cho, Minju Jo, Haksoo Lim, Kookjin Lee, Dongeun Lee, Sanghyun Hong, and Noseong Park
International Conference on Machine Learning (ICML). 2024. [Oral] Conf
When Do 'More Contexts' Help with Sarcasm Recognition?
Ojas Nimase and Sanghyun Hong
Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING). 2024. Conf
Extension of Physics-informed Neural Networks to Solving Parameterized PDEs
Woojin Cho, Minju Jo, Haksoo Lim, Kookjin Lee, Dongeun Lee, Sanghyun Hong, and Noseong Park
ICLR Workshop on AI4DifferentialEquations in Science (AI4DiffEqtnsInSci). 2024. Workshop
PAC-FNO: Parallel-Structured All-Component Fourier Neural Operators for Recognizing Low-Quality Images
Jinsung Jeon, Hyundong Jin, Jonghyun Choi, Sanghyun Hong, Dongeun Lee, Kookjin Lee, and Noseong Park
International Conference on Learning Representations (ICLR). 2024. Conf
Operator-learning-inspired Modeling of Neural Ordinary Differential Equations
Woojin Cho, Seunghyeon Cho, Hyundong Jin, Jinsung Jeon, Kookjin Lee, Sanghyun Hong, Dongeun Lee, Jonghyun Choi, and Noseong Park
AAAI Conference on Artificial Intelligence (AAAI). 2024. Conf
2023
HyperNetwork Approximating Future Parameters for Time Series Forecasting under Temporal Drifts
Jaehoon Lee, Chan Kim, Gyumin Lee, Haksoo Lim, Jeongwhan Choi, Kookjin Lee, Dongeun Lee, Sanghyun Hong, and Noseong Park
NeurIPS Workshop on Distribution Shifts (DistShift). 2023. Workshop
BERT Lost Patience Won't Be Robust to Adversarial Slowdown
Zachary Coalson, Gabriel Ritter, Rakesh Bobba, and Sanghyun Hong
Advances in Neural Information Processing Systems (NeurIPS). 2023. Conf
Learning Unforeseen Robustness from Out-of-distribution Data Using Equivariant Domain Translator
Sicheng Zhu, Bang An, Furong Huang, and Sanghyun Hong
International Conference on Machine Learning (ICML). 2023. Conf
Perceptual Pat: A Virtual Human Visual System for Iterative Visualization Design
Sungbok Shin, Sanghyun Hong, and Niklas Elmqvist
ACM Conference on Human Factors in Computing Systems (CHI). 2023. Conf
Publishing Efficient On-device Models Increases Adversarial Vulnerability
Sanghyun Hong, Nicholas Carlini, and Alexey Kurakin
IEEE Conference on Secure and Trustworthy Machine Learning (SaTML). 2023. Conf
2022
Will SOC Telemetry Data Improve Predictive Models of User Riskiness? A Work in Progress
Michael Curry, Byron Marshall, Forough Shadbad, and Sanghyun Hong
AIS SIGSEC Workshop on Information Security and Privacy (WISP). 2022. Workshop
Handcrafted Backdoors in Deep Neural Networks
Sanghyun Hong, Nicholas Carlini, and Alexey Kurakin
Advances in Neural Information Processing Systems (NeurIPS). 2022. [Oral] Conf
A Scanner Deeply: Predicting Gaze Heatmaps on Visualizations Using Crowdsourced Eye Movement Data
Sungbok Shin, Sunghyo Chung, Sanghyun Hong, and Niklas Elmqvist
IEEE Transactions on Visualization and Computer Graphics (IEEE VIS). 2022. Conf
Truth Serum: Poisoning Machine Learning Models to Reveal Their Secrets
Florian Tramèr, Reza Shokri, Ayrton San Joaquin, Hoang Le, Matthew Jagielski, Sanghyun Hong, and Nicholas Carlini
ACM Conference on Computer and Communications Security (CCS). 2022. Conf
AdamNODEs: When Neural ODE Meets Adaptive Moment Estimation
Seunghyeon Cho, Sanghyun Hong, Kookjin Lee, and Noseong Park
ICML Workshop on Continuous-Time Methods for Machine Learning. 2022. Workshop
Improving Cross-Platform Binary Analysis Using Representation Learning via Graph Alignment
Geunwoo Kim, Sanghyun Hong, Michael Franz, and Dokyung Song
ACM SIGSOFT International Symposium on Software Testing and Analysis (ISSTA). 2022. Conf
Data Poisoning Won't Save You From Facial Recognition
Evani Radiya-Dixit, Sanghyun Hong, Nicholas Carlini, and Florian Tramer
International Conference on Learning Representations (ICLR). 2022. Conf
2021
Qu-ANTI-zation: Exploiting Neural Network Quantization for Achieving Adversarial Outcomes
Sanghyun Hong, Michael-Andrei Panaitescu-Liess, Yigitcan Kaya, and Tudor Dumitraș
Advances in Neural Information Processing Systems (NeurIPS). 2021. Conf
A Panda? No, It's a Sloth: Slowdown Attacks on Adaptive Multi-Exit Neural Network Inference
Sanghyun Hong*, Yigitcan Kaya*, Ionuţ-Vlad Modoranu, and Tudor Dumitraș (*equal contribution)
International Conference on Learning Representations (ICLR). 2021. [Spotlight] Conf
Certified Malware in South Korea: A Localized Study of Breaches of Trust in Code-Signing PKI Ecosystem
Bumjun Kwon, Sanghyun Hong, Yuseok Jeon, and Doowon Kim
International Conference on Information and Communications Security (ICICS). 2021. Conf
A Sound Mind in a Vulnerable Body: Practical Hardware Attacks on Deep Learning
Sanghyun Hong
USENIX Enigma (Enigma). 2021. Conf
2020
How to 0wn NAS in Your Spare Time
Sanghyun Hong, Michael Davinroy, Yigitcan Kaya, Dana Dachman-Soled, and Tudor Dumitraș
International Conference on Learning Representations (ICLR). 2020. Conf
2019
Terminal Brain Damage: Exposing the Graceless Degradation in Deep Neural Networks Under Hardware Fault Attacks
Sanghyun Hong, Pietro Frigo, Yigitcan Kaya, Cristiano Giuffrida, and Tudor Dumitraș
USENIX Security Symposium (USENIX Security). 2019. Conf
Shallow-Deep Networks: Understanding and Mitigating Network Overthinking
Yigitcan Kaya, Sanghyun Hong, and Tudor Dumitraș
International Conference on Machine Learning (ICML). 2019. Conf
Poster: On the Feasibility of Training Neural Networks with Visibly Watermarked Dataset
Sanghyun Hong, Tae-hoon Kim, Tudor Dumitraș, and Jonghyun Choi
Network and Distributed System Security Symposium (NDSS). 2019. Workshop
Peek-a-Boo: Inferring Program Behaviors in a Virtualized Infrastructure without Introspection
Sanghyun Hong, Alina Nicolae, Abhinav Srivastava, and Tudor Dumitraș
Computer & Security (COSE). 2019. Journal
2018
Go Serverless: Securing Cloud via Serverless Design Patterns
Sanghyun Hong, Abhinav Srivastava, William Shambrook, and Tudor Dumitraș
USENIX Workshop on Hot Topics in Cloud Computing (HotCloud). 2018. Workshop
PAGE: Answering Graph Pattern Queries via Knowledge Graph Embedding
Sanghyun Hong, Noseong Park, Tanmoy Chakraborty, Hyunjoong Kang, and Soonhyun Kwon
International Conference on Big Data (Big Data). 2018. Conf
On Integrating Knowledge Graph Embedding into SPARQL Query Processing
Soonhyun Kwon, Hyunjoong Kang, Sanghyun Hong, Kookjin Lee, and Noseong Park
IEEE International Conference on Web Services (ICWS). 2018. Conf
2017
SENA: Preserving Social Structure for Network Embedding
Sanghyun Hong*, Tanmoy Chakraborty*, Sungjin Ahn, Ghaith Husari, and Noseong Park (*equal contribution)
ACM Conference on Hypertext and Social Media (ACM HT). 2017. Conf
Summoning Demons: The Pursuit of Exploitable Bugs in Machine Learning
Rock Stevens, Octavian Suciu, Andrew Ruef, Sanghyun Hong, Michael Hicks, and Tudor Dumitraș
NeurIPS Workshop on Reliable Machine Learning in the Wild. 2017. Workshop