相关论文: Causal-Guided Detoxify Backdoor Attack of Open-Wei…

Large language models (LLMs) have demonstrated superior performance compared to previous methods on various tasks, and often serve as the foundation models for many researches and services. However, the untrustworthy third-party LLMs may…

Offline Reinforcement Learning (RL) enables policy optimization from static datasets but is inherently vulnerable to backdoor attacks. Existing attack strategies typically struggle against safety-constrained algorithms (e.g., CQL) due to…

LoRA adapters let users fine-tune large language models (LLMs) efficiently. However, LoRA adapters are shared through open repositories like Hugging Face Hub \citep{huggingface_hub_docs}, making them vulnerable to backdoor attacks. Current…

Low rank adaptation (LoRA) has emerged as a prominent technique for fine-tuning large language models (LLMs) thanks to its superb efficiency gains over previous methods. While extensive studies have examined the performance and structural…

Low-Rank Adaptation (LoRA) has emerged as a leading technique for efficiently fine-tuning text-to-image diffusion models, and its widespread adoption on open-source platforms has fostered a vibrant culture of model sharing and…

Low-Rank Adaptation (LoRA) is widely used for parameter-efficient fine-tuning of large language models, but it is notably ineffective at removing backdoor behaviors from poisoned pretrained models when fine-tuning on clean dataset. Contrary…

We show that LoRA adapters, the dominant distribution format for fine-tuned LLMs, can be reliably backdoored through training data poisoning while preserving baseline task performance. On a Qwen 2.5 1.5B prompt-injection classifier, a small…

Large models adaptation through Federated Learning (FL) addresses a wide range of use cases and is enabled by Parameter-Efficient Fine-Tuning techniques such as Low-Rank Adaptation (LoRA). However, this distributed learning paradigm faces…

Model merging is an emerging technique that integrates multiple models fine-tuned on different tasks to create a versatile model that excels in multiple domains. This scheme, in the meantime, may open up backdoor attack opportunities where…

By injecting a small number of poisoned samples into the training set, backdoor attacks aim to make the victim model produce designed outputs on any input injected with pre-designed backdoors. In order to achieve a high attack success rate…

Self-Supervised Learning (SSL) has emerged as a significant paradigm in representation learning thanks to its ability to learn without extensive labeled data, its strong generalization capabilities, and its potential for privacy…

Recent advances in vision-language-action (VLA) models have greatly improved embodied AI, enabling robots to follow natural language instructions and perform diverse tasks. However, their reliance on uncurated training datasets raises…

Model quantization is a popular technique for deploying deep learning models on resource-constrained environments. However, it may also introduce previously overlooked security risks. In this work, we present QuRA, a novel backdoor attack…

Deep speech classification tasks, including keyword spotting and speaker verification, are vital in speech-based human-computer interaction. Recently, the security of these technologies has been revealed to be susceptible to backdoor…

The implications of backdoor attacks on English-centric large language models (LLMs) have been widely examined - such attacks can be achieved by embedding malicious behaviors during training and activated under specific conditions that…

Generative large language models are crucial in natural language processing, but they are vulnerable to backdoor attacks, where subtle triggers compromise their behavior. Although backdoor attacks against LLMs are constantly emerging,…

Low-Rank Adaptation (LoRA) has become a popular solution for fine-tuning large language models (LLMs) in federated settings, dramatically reducing update costs by introducing trainable low-rank matrices. However, when integrated with…

Deep Neural Networks (DNNs) are shown to be vulnerable to backdoor poisoning attacks, with most research focusing on digital triggers -- artificial patterns added to test-time inputs to induce targeted misclassification. Physical triggers,…

Recent studies have widely investigated backdoor attacks on Large Language Models (LLMs) by inserting harmful question-answer (QA) pairs into their training data. However, we revisit existing attacks and identify two critical limitations:…

Safety alignment mechanisms in Large Language Models (LLMs) often operate as latent internal states, obscuring the model's inherent capabilities. Building on this observation, we model the safety mechanism as an unobserved confounder from a…