arXiv subjects

Chia-Mu Yu

Publications and source records attributed to Chia-Mu Yu.

At least 19 recordsLinked to original sources

arXiv2026

Daydreaming: Stealing Hidden Agent Skills through Black-Box Task Interaction

Agent skills bundle instructions, reference data, and executable helpers that let a general agent perform specialized tasks. Hosted providers can keep these files secret while selling access to task results, making the skill itself a valuable target. Existing disclosure defenses can block requests that ask for the skill or reproduce its text, but they cannot block customers from submitting the ordinary tasks the service is built to complete. We present Daydreaming, an execution-only attack that steals a multi-file skill through black-box task interactions. The victim is never asked to reveal the skill or grade a reconstruction. Instead, Daydreaming adaptively creates crafted tasks whose results distinguish possible hidden behaviors. It tests individual behaviors, uses attacker-controlled shadow agents to choose a design, and completes each file using stored victim results and local execution checks. We formalize three nested threat levels of access as Differential, Trace, and Output, and focus on Output, where the attacker sees only the final response and returned files. Across 7 skills and 4 victim models, Daydreaming recovers 86.8% of the original skill's capability at Output, outperforming SigLeak by almost 4x. It produces installable skills using a median of 32 victim calls per skill even with disclosure defenses enabled. These results show that hiding skill files and filtering direct disclosure do not, by themselves, prevent functional reconstruction through normal use.

cs.CR↗

arXiv2026

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

Post-Quantum Cryptography (PQC) is increasingly being integrated into TLS 1.3 to enhance resilience against quantum-enabled attacks. However, the additional computational and communication overhead introduced by PQC primitives during the handshake phase may also amplify the impact of TLS handshake exhaustion attacks, leading to more severe Distributed Denial-of-Service (DDoS) threats. In this study, we establish an empirical testbed consisting of one PQC-enabled TLS server and ten attacking nodes, generating over 16.5 GB of mixed traffic data that includes both legitimate browsing behavior and high-intensity handshake exhaustion attacks. Experimental results show that PQC-TLS can prolong periods of sustained high CPU utilization on the server by up to 88 times, significantly amplifying the effectiveness of such attacks. Furthermore, we evaluate state-of-the-art deep learning-based Intrusion Detection Systems (IDS) and observe a substantial decline in attack detection performance under PQC traffic conditions. In particular, exosphere achieves only around 50% recall, while HyperVision's AU-ROC degrades to near-random levels (0.49), revealing critical detection blind spots in existing IDS when operating in PQC environments. The main contributions of this work are threefold: (1) we systematically quantify and analyze the root causes of IDS detection blind spots in PQC settings; (2) we publicly release a comprehensive PQC-DDoS hybrid traffic dataset, including precise attack timestamps and server-side resource monitoring data; and (3) we open-source all experimental code and AWS deployment scripts, enabling a fully reproducible cloud-based testing environment. These resources aim to support both academia and industry in developing next-generation PQC-aware intrusion detection systems.

cs.CR↗

arXiv2026

Tracing Target Answers in Poisoned Retrieval Corpora via Token Influence Attribution

Retrieval-Augmented Generation (RAG) systems are vulnerable to corpus poisoning attacks that manipulate model outputs through malicious retrieved documents. Existing detection methods typically rely on auxiliary classifiers or additional LLM-based verification, introducing substantial computational overhead. We present TRACE, a lightweight detection framework that identifies poisoning attacks by tracing answer-related tokens through token influence attribution. TRACE first discovers recurrent high-influence keywords across retrieved documents and then performs a secondary verification to confirm their influence on model predictions. Experiments on three QA benchmarks and six LLMs demonstrate strong detection performance while simultaneously uncovering attacker-specified target answers.

cs.CR↗

arXiv2026

RAS: Measuring LLM Safety Through Refusal Alignment

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 policy. Although useful, output-level evaluation is expensive, sensitive to judge choice, and easily tied to fixed question banks. We propose **SafeVec**, a white-box evaluation procedure that measures safety from internal representations rather than generated answers. **SafeVec** first extracts layer-wise refusal directions from a safety-aligned reference model, then selects stable layer windows where safe and unsafe behaviors are separable, and finally scores a target model by measuring whether its hidden states align with these refusal directions under unsafe and jailbreak prompts. The resulting metric, **RAS** (**R**efusal **A**lignment **S**core), maps representation-level refusal alignment to a calibrated 0-100 safety score. Across `Llama`, `Gemma`, and `Qwen` model families, RAS separates aligned models from uncensored and abliterated variants, tracks output-level attack success rate, and is substantially faster than judge-based evaluation. These results suggest that refusal alignment provides a compact and efficient signal for white-box LLM safety evaluation.

cs.CR↗

arXiv2026

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

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 revealed a new security threat known as Quantization-Conditioned Backdoors (QCBs), where a model behaves normally in full precision but activates malicious behavior only after quantization. Existing defenses typically modify quantization procedures or correct activation statistics, often introducing additional computational overhead or relying on specific quantization settings. Here, we present QVec, a parameter-space perspective for defending against QCBs. We observe that the weight difference between a full-precision model and its quantized counterpart encodes a structured behavioral shift, which can be interpreted as a malicious task vector rather than random quantization noise. Based on this insight, QVec counteracts this malicious direction through controlled parameter correction prior to deployment. QVec requires no retraining, no trigger samples, and only a single quantization pass to estimate the parameter shift, together with a lightweight hyperparameter search. Extensive experiments across image classification benchmarks and multiple Large Language Model (LLM) attack scenarios demonstrate that QVec consistently suppresses backdoor activation while preserving clean performance.

cs.CR↗

arXiv2026

PhantomSkill: Malicious Code Injection in Agent Skill Ecosystems

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 PhantomSkill, an attack framework that hides malicious behavior in a skill's auxiliary resources rather than in its textual description. Its core technique, VulMask, rewrites overt malicious scripts into vulnerability-shaped implementations whose malicious behavior is activated only under attacker-controlled trigger conditions. This design shifts the visible signal from explicit malicious intent to ordinary-looking insecure code. Across representative host skills, attack goals, coding agents, generation models, and automated reviewers, VulMask preserves benign utility while reducing warning and malware-level detection compared with overt malicious scripts. Our results show that skill ecosystems require resource-level vetting, execution-time containment, and security policies that treat exploitable vulnerabilities in agent skills as potential malicious payloads.

cs.CR↗

arXiv2026

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

Code large language models increasingly retrieve external code context from repositories, documentation, issue threads, and coding-agent environments, creating an indirect prompt-injection surface where attackers hide instructions in comments, strings, identifiers, or decoy code. We propose CodeSentinel, a three-layer inference-time sanitizer. It uses Tree-sitter to extract high-risk model-facing CST nodes, then combines syntax-guided pre-filtering, CST-guided Dynamic Min-K\% scoring, and node perturbation analysis to detect adversarial and natural-looking semantic triggers. Detected nodes are removed or neutralized before reaching the downstream Code LLM. Across six recent attack families, \CodeSentinel achieves 0.80 average node-level F1, outperforming CodeGarrison, DePA, and KillBadCode.

cs.CR↗

arXiv2026

WebMCP Tool Surface Poisoning: Runtime Manipulation Attacks on LLM Agents

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 dynamic exposure of agent-accessible tools in WebMCP expands the attack surface of web sessions, especially when third-party scripts are involved. In this study, we identify a new potential threat, termed Mid-Session Tool Injection (MSTI), in which attackers leverage third-party scripts to inject malicious tools during an active session. To better characterize this threat, we classify MSTI based on the stage and target of manipulation, distinguishing between Tool Hijacking and Tool Framing. Tool Hijacking modifies the set of tools visible to the agent through mechanisms such as the AbortSignal API or race conditions during tool registration. In contrast, Tool Framing influences the agent's perception of tool roles through metadata fields such as tool name, description, readOnlyHint, and inputSchema. Our implementation demonstrates that both Tool Hijacking and Tool Framing can successfully disrupt the intended functionality of WebMCP. Based on these results, we outline potential mitigation directions and provide security design recommendations for WebMCP, including binding tool identity to its origin, ensuring lifecycle consistency, enforcing data boundaries for third-party tools, and maintaining traceable logs of tool registration and invocation. These findings indicate that MSTI arises from WebMCP's unique tool lifecycle and structured metadata, making the tool surface itself an emerging security concern.

cs.CR↗

arXiv2026

Hidden Thoughts Are Not Secret: Reasoning Trace Exposure in LLMs

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 distill reasoning behavior from stronger teacher models into weaker student models. The value of capability transfer has motivated many deployed systems with reasoning models to hide raw internal traces and expose at most summaries and answers to users. As a result, we ask whether such interface-level trace hiding prevents users from obtaining useful reasoning supervision through prompting. We study this question with Reasoning Exposure Prompting (REP), a lightweight in-context elicitation method that uses shadow-model-generated demonstrations wrapped in auxiliary code-like formats to raise user-visible reasoning traces from a victim model. Across the common reasoning dataset, different victim models, and different student model distillation, REP substantially increases similarity between exposed and REP-conditioned internal traces while preserving useful reasoning signals.

cs.AI↗

arXiv2026

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

LLM-powered coding agents increasingly participate in software development workflows by generating code, selecting dependencies, and producing package installation commands. This creates a new software supply chain risk: when an agent hallucinates a non-existent package, an attacker may register the hallucinated name and later compromise users who install it. Existing package hallucination attacks and defenses primarily focus on naturally occurring hallucinations, targeted dependency steering, or post-hoc package validation. In this paper, we introduce \emph{Neutral Prompting Attack} (NPA), a highly stealthy attack paradigm in which semantically benign instructions, such as encouraging imagination and exhaustiveness, increase package hallucination propensity without containing explicit malicious intent. Unlike targeted dependency steering, NPA does not specify an attacker-chosen package. Instead, it shifts the model's dependency generation behavior toward more speculative package names. We evaluate NPA across multiple coding-oriented LLMs and package hallucination benchmarks. Our results show that NPA increases both \emph{Hallucination ASR} and \emph{Pip Install ASR}, changes the distribution of hallucinated package names, and evades existing static-analysis, LLM-based, and agent-based Skill defenses. These findings reveal that harmless-looking prompts can covertly manipulate hallucination behavior and create downstream software supply chain risks.

cs.CR↗

arXiv2026

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

LLM-powered coding agents increasingly make software supply chain decisions. They generate imports, recommend packages, and write installation commands. Prior work showed that these systems can hallucinate non-existent package names, which attackers may register as malicious packages. In this paper, we show that this risk is not only a passive model failure. It can be actively induced through the persistent Skill artifact. We introduce Dependency Steering, an attack paradigm in which a malicious Skill biases a coding agent toward an attacker-controlled package during benign coding tasks. The attack does not require modifying model weights, training data, or user prompts. To construct realistic attacks, we design a Skill-level optimization method that searches for localized semantic edits that preserve the apparent purpose of the original Skill while increasing targeted package generation. Across multiple coding-oriented LLMs and programming benchmarks, Dependency Steering achieves high targeted hallucination rates, transfers across models and task domains, and remains difficult for evaluated Skill scanners and LLM-based auditors to detect. Our results show that persistent agent instructions form an underexplored software supply chain attack surface.

cs.CR↗

arXiv2026

IU: Imperceptible Universal Backdoor Attack

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 easier to detect and less practical at scale. In this work, we introduce a novel imperceptible universal backdoor attack that simultaneously controls all target classes with minimal poisoning while preserving stealth. Our key idea is to leverage graph convolutional networks (GCNs) to model inter-class relationships and generate class-specific perturbations that are both effective and visually invisible. The proposed framework optimizes a dual-objective loss that balances stealthiness (measured by perceptual similarity metrics such as PSNR) and attack success rate (ASR), enabling scalable, multi-target backdoor injection. Extensive experiments on ImageNet-1K with ResNet architectures demonstrate that our method achieves high ASR (up to 91.3%) under poisoning rates as low as 0.16%, while maintaining benign accuracy and evading state-of-the-art defenses. These results highlight the emerging risks of invisible universal backdoors and call for more robust detection and mitigation strategies.

cs.CR↗

arXiv2025

Defending Unauthorized Model Merging via Dual-Stage Weight Protection

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 new multi-capability model without authorization. Such unauthorized model merging not only violates intellectual property rights but also undermines model ownership and accountability. To address this issue, we present MergeGuard, a proactive dual-stage weight protection framework that disrupts merging compatibility while maintaining task fidelity. In the first stage, we redistribute task-relevant information across layers via L2-regularized optimization, ensuring that important gradients are evenly dispersed. In the second stage, we inject structured perturbations to misalign task subspaces, breaking curvature compatibility in the loss landscape. Together, these stages reshape the model's parameter geometry such that merged models collapse into destructive interference while the protected model remains fully functional. Extensive experiments on both vision (ViT-L-14) and language (Llama2, Gemma2, Mistral) models demonstrate that MergeGuard reduces merged model accuracy by up to 90% with less than 1.5% performance loss on the protected model.

cs.CV↗

arXiv2025

Model Reprogramming Demystified: A Neural Tangent Kernel Perspective

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 solution to the challenges of training large models for diverse tasks. Despite its empirical success across various domains such as computer vision and time-series forecasting, the theoretical foundations of MR remain underexplored. In this paper, we present a comprehensive theoretical analysis of MR through the lens of the Neural Tangent Kernel (NTK) framework. We demonstrate that the success of MR is governed by the eigenvalue spectrum of the NTK matrix on the target dataset and establish the critical role of the source model's effectiveness in determining reprogramming outcomes. Our contributions include a novel theoretical framework for MR, insights into the relationship between source and target models, and extensive experiments validating our findings.

cs.LG↗

arXiv2025

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

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 types (e.g., categorical, numerical, and graph data) and application settings (e.g., static and streaming). Recent findings indicate that LDP protocols can be easily disrupted by poisoning or manipulation attacks, which leverage injected/corrupted fake users to send crafted data conforming to the LDP reports. However, current attacks primarily target static protocols, neglecting the security of LDP protocols in the streaming settings. Our research fills the gap by developing novel fine-grained manipulation attacks to LDP protocols for data streams. By reviewing the attack surfaces in existing algorithms, We introduce a unified attack framework with composable modules, which can manipulate the LDP estimated stream toward a target stream. Our attack framework can adapt to state-of-the-art streaming LDP algorithms with different analytic tasks (e.g., frequency and mean) and LDP models (event-level, user-level, w-event level). We validate our attacks theoretically and through extensive experiments on real-world datasets, and finally explore a possible defense mechanism for mitigating these attacks.

cs.CR↗

arXiv2025

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

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 synthetic data, especially for high-resolution images. On the other hand, one of the emerging techniques in parameter efficient fine-tuning (PEFT) is visual prompting (VP), which allows well-trained existing models to be reused for the purpose of adapting to subsequent downstream tasks. In this work, we explore such a phenomenon in constructing captivating generative models with DP constraints. We show that VP in conjunction with DP-NTK, a DP generator that exploits the power of the neural tangent kernel (NTK) in training DP generative models, achieves a significant performance boost, particularly for high-resolution image datasets, with accuracy improving from 0.644$\pm$0.044 to 0.769. Lastly, we perform ablation studies on the effect of different parameters that influence the overall performance of VP-NTK. Our work demonstrates a promising step forward in improving the utility of DP synthetic data, particularly for high-resolution images.

cs.CV↗

arXiv2025

Poisoning Attacks to Local Differential Privacy Protocols for Trajectory Data

Trajectory data, which tracks movements through geographic locations, is crucial for improving real-world applications. However, collecting such sensitive data raises considerable privacy concerns. Local differential privacy (LDP) offers a solution by allowing individuals to locally perturb their trajectory data before sharing it. Despite its privacy benefits, LDP protocols are vulnerable to data poisoning attacks, where attackers inject fake data to manipulate aggregated results. In this work, we make the first attempt to analyze vulnerabilities in several representative LDP trajectory protocols. We propose \textsc{TraP}, a heuristic algorithm for data \underline{P}oisoning attacks using a prefix-suffix method to optimize fake \underline{Tra}jectory selection, significantly reducing computational complexity. Our experimental results demonstrate that our attack can substantially increase target pattern occurrences in the perturbed trajectory dataset with few fake users. This study underscores the urgent need for robust defenses and better protocol designs to safeguard LDP trajectory data against malicious manipulation.

cs.CR↗

arXiv2025

Data Poisoning Attacks to Locally Differentially Private Range Query Protocols

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 vulnerable to data poisoning attacks, where malicious users manipulate their reported data to distort aggregated results. In this work, we present the first study on data poisoning attacks targeting LDP range query protocols, focusing on both tree-based and grid-based approaches. We identify three key challenges in executing such attacks, including crafting consistent and effective fake data, maintaining data consistency across levels or grids, and preventing server detection. To address the first two challenges, we propose novel attack methods that are provably optimal, including a tree-based attack and a grid-based attack, designed to manipulate range query results with high effectiveness. \textbf{Our key finding is that the common post-processing procedure, Norm-Sub, in LDP range query protocols can help the attacker massively amplify their attack effectiveness.} In addition, we study a potential countermeasure, but also propose an adaptive attack capable of evading this defense to address the third challenge. We evaluate our methods through theoretical analysis and extensive experiments on synthetic and real-world datasets. Our results show that the proposed attacks can significantly amplify estimations for arbitrary range queries by manipulating a small fraction of users, providing 5-10x more influence than a normal user to the estimation.

cs.CR↗