papers

Publications (53)

cs.CR2021

Real-World Adversarial Examples involving Makeup Application

Chang-Sheng Lin, Chia-Yi Hsu, Pin-Yu Chen +1

Deep neural networks have developed rapidly and have achieved outstanding performance in several tasks, such as image classification and natural language processing. However, recen…

cs.LG2025

Model Reprogramming Demystified: A Neural Tangent Kernel Perspective

Ming-Yu Chung, Jiashuo Fan, Hancheng Ye +5

Model Reprogramming (MR) is a resource-efficient framework that adapts large pre-trained models to new tasks with minimal additional parameters and data, offering a promising solut…

cs.LG2024

Ring-A-Bell! How Reliable are Concept Removal Methods for Diffusion Models?

Yu-Lin Tsai, Chia-Yi Hsu, Chulin Xie +6

Diffusion models for text-to-image (T2I) synthesis, such as Stable Diffusion (SD), have recently demonstrated exceptional capabilities for generating high-quality content. However,…

cs.LG2022

CAFE: Catastrophic Data Leakage in Vertical Federated Learning

Xiao Jin, Pin-Yu Chen, Chia-Yi Hsu +2

Recent studies show that private training data can be leaked through the gradients sharing mechanism deployed in distributed machine learning systems, such as federated learning (F…

cs.CV2021

Perceptual Indistinguishability-Net (PI-Net): Facial Image Obfuscation with Manipulable Semantics

Jia-Wei Chen, Li-Ju Chen, Chia-Mu Yu +1

With the growing use of camera devices, the industry has many image datasets that provide more opportunities for collaboration between the machine learning community and industry.…

cs.LG2025

Rethinking Backdoor Attacks on Dataset Distillation: A Kernel Method Perspective

Ming-Yu Chung, Sheng-Yen Chou, Chia-Mu Yu +3

Dataset distillation offers a potential means to enhance data efficiency in deep learning. Recent studies have shown its ability to counteract backdoor risks present in original tr…

cs.CR2026

Quantization as a Malicious Task: Removing Quantization-Conditioned Backdoors via Task Arithmetic

Kaihsun Yang, Min-Yan Tsai, Chia-Mu Yu

Model quantization is widely adopted to reduce memory usage and inference cost when deploying deep neural networks on resource-constrained devices. However, recent studies have rev…

cs.CL2025

Beyond Natural Language Perplexity: Detecting Dead Code Poisoning in Code Generation Datasets

Chi-Chien Tsai, Chia-Mu Yu, Ying-Dar Lin +2

The increasing adoption of large language models (LLMs) for code-related tasks has raised concerns about the security of their training datasets. One critical threat is dead code p…

cs.CV2024

Differentially Private Fine-Tuning of Diffusion Models

Yu-Lin Tsai, Yizhe Li, Zekai Chen +4

The integration of Differential Privacy (DP) with diffusion models (DMs) presents a promising yet challenging frontier, particularly due to the substantial memorization capabilitie…

cs.CV2018

On The Utility of Conditional Generation Based Mutual Information for Characterizing Adversarial Subspaces

Chia-Yi Hsu, Pei-Hsuan Lu, Pin-Yu Chen +1

Recent studies have found that deep learning systems are vulnerable to adversarial examples; e.g., visually unrecognizable adversarial images can easily be crafted to result in mis…

cs.LG2023

Exploring the Benefits of Differentially Private Pre-training and Parameter-Efficient Fine-tuning for Table Transformers

Xilong Wang, Chia-Mu Yu, Pin-Yu Chen

For machine learning with tabular data, Table Transformer (TabTransformer) is a state-of-the-art neural network model, while Differential Privacy (DP) is an essential component to…

cs.DC2020

DPCrowd: Privacy-preserving and Communication-efficient Decentralized Statistical Estimation for Real-time Crowd-sourced Data

Xuebin Ren, Chia-Mu Yu, Wei Yu +3

In Internet of Things (IoT) driven smart-world systems, real-time crowd-sourced databases from multiple distributed servers can be aggregated to extract dynamic statistics from a l…

cs.CR2026

Harmless Yet Harmful: Neutral Prompting Attacks for Stealthy Hallucination Steering in Agent Skills

Chia-Yi Hsu, Chia-Mu Yu, Chun-Ying Huang +1

LLM-powered coding agents increasingly participate in software development workflows by generating code, selecting dependencies, and producing package installation commands. This c…

cs.CR2025

Fine-grained Manipulation Attacks to Local Differential Privacy Protocols for Data Streams

Xinyu Li, Xuebin Ren, Shusen Yang +2

Local Differential Privacy (LDP) enables massive data collection and analysis while protecting end users' privacy against untrusted aggregators. It has been applied to various data…

math.ST2019

Locally Differentially Private Minimum Finding

Kazuto Fukuchi, Chia-Mu Yu, Arashi Haishima +1

We investigate a problem of finding the minimum, in which each user has a real value and we want to estimate the minimum of these values under the local differential privacy constr…

cs.CR2026

CodeSentinel: A Three-Layer Defense Against Indirect Prompt Injection in Code Contexts

Po-Han Cheng, Chia-Mu Yu, Ying-Dar Lin +2

Code large language models increasingly retrieve external code context from repositories, documentation, issue threads, and coding-agent environments, creating an indirect prompt-i…

cs.CR2026

WebMCP Tool Surface Poisoning: Runtime Manipulation Attacks on LLM Agents

Lin-Fa Lee, Yi-Yu Chang, Chia-Mu Yu +1

WebMCP is a newly emerging protocol that enables websites to expose tools directly to AI agents, bypassing traditional user interfaces and introducing new security risks. The dynam…

cs.CV2023

DPAF: Image Synthesis via Differentially Private Aggregation in Forward Phase

Chih-Hsun Lin, Chia-Yi Hsu, Chia-Mu Yu +2

Differentially private synthetic data is a promising alternative for sensitive data release. Many differentially private generative models have been proposed in the literature. Unf…

cs.LG2018

On the Limitation of Local Intrinsic Dimensionality for Characterizing the Subspaces of Adversarial Examples

Pei-Hsuan Lu, Pin-Yu Chen, Chia-Mu Yu

Understanding and characterizing the subspaces of adversarial examples aid in studying the robustness of deep neural networks (DNNs) to adversarial perturbations. Very recently, Ma…

cs.LG2021

Non-Singular Adversarial Robustness of Neural Networks

Yu-Lin Tsai, Chia-Yi Hsu, Chia-Mu Yu +1

Adversarial robustness has become an emerging challenge for neural network owing to its over-sensitivity to small input perturbations. While being critical, we argue that solving t…

cs.CV2025

BadVim: Unveiling Backdoor Threats in Visual State Space Model

Cheng-Yi Lee, Yu-Hsuan Chiang, Zhong-You Wu +2

Visual State Space Models (VSSM) have shown remarkable performance in various computer vision tasks. However, backdoor attacks pose significant security challenges, causing comprom…

cs.CV2025

Prompting the Unseen: Detecting Hidden Backdoors in Black-Box Models

Zi-Xuan Huang, Jia-Wei Chen, Zhi-Peng Zhang +1

Visual prompting (VP) is a new technique that adapts well-trained frozen models for source domain tasks to target domain tasks. This study examines VP's benefits for black-box mode…

cs.LG2021

Meta Adversarial Perturbations

Chia-Hung Yuan, Pin-Yu Chen, Chia-Mu Yu

A plethora of attack methods have been proposed to generate adversarial examples, among which the iterative methods have been demonstrated the ability to find a strong attack. Howe…

cs.CV2019

Detecting Deepfake-Forged Contents with Separable Convolutional Neural Network and Image Segmentation

Chia-Mu Yu, Ching-Tang Chang, Yen-Wu Ti

Recent advances in AI technology have made the forgery of digital images and videos easier, and it has become significantly more difficult to identify such forgeries. These forgeri…

cs.CR2026

IU: Imperceptible Universal Backdoor Attack

Hsin Lin, Yan-Lun Chen, Ren-Hung Hwang +1

Backdoor attacks pose a critical threat to the security of deep neural networks, yet existing efforts on universal backdoors often rely on visually salient patterns, making them ea…

cs.CV2018

On the Limitation of MagNet Defense against -based Adversarial Examples

Pei-Hsuan Lu, Pin-Yu Chen, Kang-Cheng Chen +1

In recent years, defending adversarial perturbations to natural examples in order to build robust machine learning models trained by deep neural networks (DNNs) has become an emerg…

cs.CR2025

Poisoning Attacks to Local Differential Privacy Protocols for Trajectory Data

I-Jung Hsu, Chih-Hsun Lin, Chia-Mu Yu +2

Trajectory data, which tracks movements through geographic locations, is crucial for improving real-world applications. However, collecting such sensitive data raises considerable…

cs.LG2025

Safety Alignment Depth in Large Language Models: A Markov Chain Perspective

Ching-Chia Kao, Chia-Mu Yu, Chun-Shien Lu +1

Large Language Models (LLMs) are increasingly adopted in high-stakes scenarios, yet their safety mechanisms often remain fragile. Simple jailbreak prompts or even benign fine-tunin…

cs.CR2017

LoPub: High-Dimensional Crowdsourced Data Publication with Local Differential Privacy

Xuebin Ren, Chia-Mu Yu, Weiren Yu +4

High-dimensional crowdsourced data collected from a large number of users produces rich knowledge for our society. However, it also brings unprecedented privacy threats to particip…

cs.LG2025

BADTV: Unveiling Backdoor Threats in Third-Party Task Vectors

Chia-Yi Hsu, Yu-Lin Tsai, Yu Zhe +6

Task arithmetic in large-scale pre-trained models enables agile adaptation to diverse downstream tasks without extensive retraining. By leveraging task vectors (TVs), users can per…

cs.LG2025

The Great Contradiction Showdown: How Jailbreak and Stealth Wrestle in Vision-Language Models?

Ching-Chia Kao, Chia-Mu Yu, Chun-Shien Lu +1

Vision-Language Models (VLMs) have achieved remarkable performance on a variety of tasks, yet they remain vulnerable to jailbreak attacks that compromise safety and reliability. In…

cs.LG2021

Formalizing Generalization and Robustness of Neural Networks to Weight Perturbations

Yu-Lin Tsai, Chia-Yi Hsu, Chia-Mu Yu +1

Studying the sensitivity of weight perturbation in neural networks and its impacts on model performance, including generalization and robustness, is an active research topic due to…

cs.CR2026

Tracing Target Answers in Poisoned Retrieval Corpora via Token Influence Attribution

Yan-Lun Chen, Pin-Yu Chen, Chia-Mu Yu +3

Retrieval-Augmented Generation (RAG) systems are vulnerable to corpus poisoning attacks that manipulate model outputs through malicious retrieved documents. Existing detection meth…

cs.DB2020

Privacy in Data Service Composition

Mahmoud Barhamgi, Charith Perera, Chia-Mu Yu +3

In modern information systems different information features, about the same individual, are often collected and managed by autonomous data collection services that may have differ…

cs.CR2025

Data Poisoning Attacks to Locally Differentially Private Range Query Protocols

Ting-Wei Liao, Chih-Hsun Lin, Yu-Lin Tsai +5

Local Differential Privacy (LDP) has been widely adopted to protect user privacy in decentralized data collection. However, recent studies have revealed that LDP protocols are vuln…

cs.CL2025

Layer-Aware Task Arithmetic: Disentangling Task-Specific and Instruction-Following Knowledge

Yan-Lun Chen, Yi-Ru Wei, Chia-Yi Hsu +5

Large language models (LLMs) demonstrate strong task-specific capabilities through fine-tuning, but merging multiple fine-tuned models often leads to degraded performance due to ov…

cs.CR2017

Data-Driven and Deep Learning Methodology for Deceptive Advertising and Phone Scams Detection

TonTon Hsien-De Huang, Chia-Mu Yu, Hung-Yu Kao

The advance of smartphones and cellular networks boosts the need of mobile advertising and targeted marketing. However, it also triggers the unseen security threats. We found that…

cs.CR2026

On the Security Implications of PQC in TLS: Handshake Exhaustion and IDS Degradation

Lin-Fa Lee, Yi-Yu Chang, Chia-Mu Yu +1

The paper investigates how integrating post‑quantum cryptography into TLS 1.3 increases the server’s vulnerability to handshake‑exhaustion DDoS attacks and degrades the performance…

#post-quantum cryptography#tls#denial-of-service#intrusion detection
cs.IT2012

Reducing Reconciliation Communication Cost with Compressed Sensing

H. T. Kung, Chia-Mu Yu

We consider a reconciliation problem, where two hosts wish to synchronize their respective sets. Efficient solutions for minimizing the communication cost between the two hosts hav…

cs.CV2023

Exploring the Benefits of Visual Prompting in Differential Privacy

Yizhe Li, Yu-Lin Tsai, Xuebin Ren +2

Visual Prompting (VP) is an emerging and powerful technique that allows sample-efficient adaptation to downstream tasks by engineering a well-trained frozen source model. In this w…

cs.LG2025

Safe LoRA: the Silver Lining of Reducing Safety Risks when Fine-tuning Large Language Models

Chia-Yi Hsu, Yu-Lin Tsai, Chih-Hsun Lin +3

While large language models (LLMs) such as Llama-2 or GPT-4 have shown impressive zero-shot performance, fine-tuning is still necessary to enhance their performance for customized…

cs.NI2009

Secure Multidimensional Queries in Tiered Sensor Networks

Chia-Mu Yu, Chun-Shien Lu, Sy-Yen Kuo

In this paper, aiming at securing range query, top-k query, and skyline query in tiered sensor networks, we propose the Secure Range Query (SRQ), Secure Top- Query (STQ), and Se…

cs.CR2026

PhantomSkill: Malicious Code Injection in Agent Skill Ecosystems

Yu-Ting Lin, Chia-Mu Yu

Agent skills allow LLM-based coding agents to acquire domain-specific capabilities from third-party packages, but they also introduce a new supply-chain attack surface. We present…

cs.CR2026

RAS: Measuring LLM Safety Through Refusal Alignment

Chang-Chieh Huang, Yan-Lun Chen, Chia-Mu Yu +1

Safety evaluation of large language models (LLMs) is commonly performed by querying models with unsafe or jailbreak prompts and judging whether their outputs violate a safety polic…

cs.CR2026

Trust Me, Import This: Dependency Steering Attacks via Malicious Agent Skills

Yiyong Liu, Chia-Yi Hsu, Chun-Ying Huang +3

LLM-powered coding agents increasingly make software supply chain decisions. They generate imports, recommend packages, and write installation commands. Prior work showed that thes…

cs.LG2021

Adversarial Examples can be Effective Data Augmentation for Unsupervised Machine Learning

Chia-Yi Hsu, Pin-Yu Chen, Songtao Lu +2

Adversarial examples causing evasive predictions are widely used to evaluate and improve the robustness of machine learning models. However, current studies focus on supervised lea…

cs.CV2024

DiffuseKronA: A Parameter Efficient Fine-tuning Method for Personalized Diffusion Models

Shyam Marjit, Harshit Singh, Nityanand Mathur +3

In the realm of subject-driven text-to-image (T2I) generative models, recent developments like DreamBooth and BLIP-Diffusion have led to impressive results yet encounter limitation…

cs.CV2024

Defending Against Repetitive Backdoor Attacks on Semi-supervised Learning through Lens of Rate-Distortion-Perception Trade-off

Cheng-Yi Lee, Ching-Chia Kao, Cheng-Han Yeh +3

Semi-supervised learning (SSL) has achieved remarkable performance with a small fraction of labeled data by leveraging vast amounts of unlabeled data from the Internet. However, th…

quant-ph2023

Certified Robustness of Quantum Classifiers against Adversarial Examples through Quantum Noise

Jhih-Cing Huang, Yu-Lin Tsai, Chao-Han Huck Yang +4

Recently, quantum classifiers have been found to be vulnerable to adversarial attacks, in which quantum classifiers are deceived by imperceptible noises, leading to misclassificati…

cs.AI2026

Hidden Thoughts Are Not Secret: Reasoning Trace Exposure in LLMs

Yu-An Lu, Ci-Yang Tsai, Yu-Lin Tsai +2

Reasoning traces have become a valuable form of learning signals for improving and transferring the capabilities of large language models. In particular, detailed traces can help d…

cs.CV2025

VP-NTK: Exploring the Benefits of Visual Prompting in Differentially Private Data Synthesis

Chia-Yi Hsu, Jia-You Chen, Yu-Lin Tsai +4

Differentially private (DP) synthetic data has become the de facto standard for releasing sensitive data. However, many DP generative models suffer from the low utility of syntheti…

cs.NI2009

Constrained Function Based En-Route Filtering for Sensor Networks

Chia-Mu Yu, Chun-Shien Lu, Sy-Yen Kuo

Sensor networks are vulnerable to \emph{false data injection attack} and \emph{path-based DoS} (PDoS) attack. While conventional authentication schemes are insufficient for solving…

cs.CV2026

Defending Unauthorized Model Merging via Dual-Stage Weight Protection

Wei-Jia Chen, Min-Yen Tsai, Cheng-Yi Lee +1

The rapid proliferation of pretrained models and open repositories has made model merging a convenient yet risky practice, allowing free-riders to combine fine-tuned models into a…