9 citations · 25 across the 25 of their papers we have counts for
28 papers
High-quality Data Do not Mean Safe! Poisoning LLMs after Data Selection
Kaiyang Li, Jiahao Chen, Yuwen Pu +5
Safety-aligned Large Language Models remain vulnerable to fine-tuning on small sets of harmful or benign-looking samples. However, prior studies typically assume that poisoned samp…
Sleeping Secrets: How Fine-Tuning Reawakens Privacy Risks in Language Models
Jianhong Li, Jiahao Chen, Yuwen Pu +6
Beyond adapting Large Language Models (LLMs) to specialized applications, fine-tuning has recently been shown to recover private information that is no longer accessible through di…
Safety in Self-Evolving Agents: A Survey
Jiahao Chen, Zhou Feng, Oubo Ma +28
Large language models (LLMs) exhibit strong general capabilities, yet their parameters typically remain fixed after deployment, limiting learning from new interactions. In open-end…
A Finger on the Scale: Covert Policy Steering through Agentic Skills
Jiarui Li, Jiahao Chen, Chunyi Zhou +5
Reusable agent skills extend large language model (LLM) agents with task procedures, tool-use guidance, and output constraints. Yet these skills also act as externalized behavioral…
The Shape of Ownership: Verifying LLM Provenance through Semantic Structures
Zhongrui Sun, Jiahao Chen, Oubo Ma +4
As large language models (LLMs) are increasingly redistributed, adapted, and served behind opaque APIs, model ownership can no longer be established reliably by inspecting model in…
Lilith: Backdoor Generalization under Training-Inference Trigger Shift
Zhou Feng, Jiahao Chen, Chunyi Zhou +6
Machine-learning services increasingly rely on public data, third-party providers, and outsourced training, creating opportunities for data-poisoning attacks that implant persisten…
Profiling for Pennies: Unveiling the Privacy Iceberg of LLM Agents
Jiahao Chen, Qi Zhang, Ruixiao Lin +7
Large Language Models (LLMs) have revolutionized how information are collected, aggregated, and reasoned. However, this enables a novel and accessible vector of privacy intrusion:…
Unveiling the Security Risks of Federated Learning in the Wild: From Research to Practice
Jiahao Chen, Zhiming Zhao, Yuwen Pu +4
Federated learning (FL) has attracted substantial attention in both academia and industry, yet its practical security posture remains poorly understood. In particular, a large body…
The Eminence in Shadow: Exploiting Feature Boundary Ambiguity for Robust Backdoor Attacks
Zhou Feng, Jiahao Chen, Chunyi Zhou +5
Deep neural networks (DNNs) underpin critical applications yet remain vulnerable to backdoor attacks, typically reliant on heuristic brute-force methods. Despite significant empiri…
FreeTalk:A plug-and-play and black-box defense against speech synthesis attacks
Yuwen Pu, Zhou Feng, Chunyi Zhou +4
Recently, speech assistant and speech verification have been used in many fields, which brings much benefit and convenience for us. However, when we enjoy these speech applications…
Enkidu: Universal Frequential Perturbation for Real-Time Audio Privacy Protection against Voice Deepfakes
Zhou Feng, Jiahao Chen, Chunyi Zhou +4
The rapid advancement of voice deepfake technologies has raised serious concerns about user audio privacy, as attackers increasingly exploit publicly available voice data to genera…
Poison in the Well: Feature Embedding Disruption in Backdoor Attacks
Zhou Feng, Jiahao Chen, Chunyi Zhou +3
Backdoor attacks embed malicious triggers into training data, enabling attackers to manipulate neural network behavior during inference while maintaining high accuracy on benign in…
UNIDOOR: A Universal Framework for Action-Level Backdoor Attacks in Deep Reinforcement Learning
Oubo Ma, Linkang Du, Yang Dai +4
Deep reinforcement learning (DRL) is widely applied to safety-critical decision-making scenarios. However, DRL is vulnerable to backdoor attacks, especially action-level backdoors,…
CLIBE: Detecting Dynamic Backdoors in Transformer-based NLP Models
Rui Zeng, Xi Chen, Yuwen Pu +3
Backdoors can be injected into NLP models to induce misbehavior when the input text contains a specific feature, known as a trigger, which the attacker secretly selects. Unlike fix…
CAMH: Advancing Model Hijacking Attack in Machine Learning
Xing He, Jiahao Chen, Yuwen Pu +5
In the burgeoning domain of machine learning, the reliance on third-party services for model training and the adoption of pre-trained models have surged. However, this reliance int…
Enhancing Adversarial Transferability with Adversarial Weight Tuning
Jiahao Chen, Zhou Feng, Rui Zeng +6
Deep neural networks (DNNs) are vulnerable to adversarial examples (AEs) that mislead the model while appearing benign to human observers. A critical concern is the transferability…
Rethinking the Vulnerabilities of Face Recognition Systems:From a Practical Perspective
Jiahao Chen, Zhiqiang Shen, Yuwen Pu +5
Face Recognition Systems (FRS) have increasingly integrated into critical applications, including surveillance and user authentication, highlighting their pivotal role in modern se…
Dullahan: Stealthy Backdoor Attack against Without-Label-Sharing Split Learning
Yuwen Pu, Zhuoyuan Ding, Jiahao Chen +4
As a novel privacy-preserving paradigm aimed at reducing client computational costs and achieving data utility, split learning has garnered extensive attention and proliferated wid…
Mellivora Capensis: A Backdoor-Free Training Framework on the Poisoned Dataset without Auxiliary Data
Yuwen Pu, Jiahao Chen, Chunyi Zhou +4
The efficacy of deep learning models is profoundly influenced by the quality of their training data. Given the considerations of data diversity, data scale, and annotation expenses…
SUB-PLAY: Adversarial Policies against Partially Observed Multi-Agent Reinforcement Learning Systems
Oubo Ma, Yuwen Pu, Linkang Du +5
Recent advancements in multi-agent reinforcement learning (MARL) have opened up vast application prospects, such as swarm control of drones, collaborative manipulation by robotic a…
The Risk of Federated Learning to Skew Fine-Tuning Features and Underperform Out-of-Distribution Robustness
Mengyao Du, Miao Zhang, Yuwen Pu +3
To tackle the scarcity and privacy issues associated with domain-specific datasets, the integration of federated learning in conjunction with fine-tuning has emerged as a practical…
MEAOD: Model Extraction Attack against Object Detectors
Zeyu Li, Chenghui Shi, Yuwen Pu +4
The widespread use of deep learning technology across various industries has made deep neural network models highly valuable and, as a result, attractive targets for potential atta…
Improving the Robustness of Transformer-based Large Language Models with Dynamic Attention
Lujia Shen, Yuwen Pu, Shouling Ji +4
Transformer-based models, such as BERT and GPT, have been widely adopted in natural language processing (NLP) due to their exceptional performance. However, recent studies show the…
Facial Data Minimization: Shallow Model as Your Privacy Filter
Yuwen Pu, Jiahao Chen, Jiayu Pan +4
Face recognition service has been used in many fields and brings much convenience to people. However, once the user's facial data is transmitted to a service provider, the user wil…