相关论文: Exploring the Universal Vulnerability of Prompt-ba…

Prompt tuning, a recently emerging paradigm, enables the powerful vision-language pre-training models to adapt to downstream tasks in a parameter -- and data -- efficient way, by learning the ``soft prompts'' to condition frozen…

Generative deep learning (DL) models have been successfully adopted for vulnerability patching. However, such models require the availability of a large dataset of patches to learn from. To overcome this issue, researchers have proposed to…

Software, while beneficial, poses potential cybersecurity risks due to inherent vulnerabilities. Detecting these vulnerabilities is crucial, and deep learning has shown promise as an effective tool for this task due to its ability to…

Backdoor attack intends to inject hidden backdoor into the deep neural networks (DNNs), such that the prediction of infected models will be maliciously changed if the hidden backdoor is activated by the attacker-defined trigger. Currently,…

Deep neural networks are vulnerable to adversarial attacks, such as backdoor attacks in which a malicious adversary compromises a model during training such that specific behaviour can be triggered at test time by attaching a specific word…

Prompt-based adversarial attacks have become an effective means to assess the robustness of large language models (LLMs). However, existing approaches often treat prompts as monolithic text, overlooking their structural…

While pre-trained Vision-Language Models (VLMs) such as CLIP exhibit impressive representational capabilities for multimodal data, recent studies have revealed their vulnerability to backdoor attacks. To alleviate the threat, existing…

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…

The increasing demand for customized Large Language Models (LLMs) has led to the development of solutions like GPTs. These solutions facilitate tailored LLM creation via natural language prompts without coding. However, the trustworthiness…

We introduce Progressive Prompts - a simple and efficient approach for continual learning in language models. Our method allows forward transfer and resists catastrophic forgetting, without relying on data replay or a large number of…

As the capabilities of pre-trained large language models (LLMs) continue to advance, the "pre-train and fine-tune" paradigm has become increasingly mainstream, leading to the development of various fine-tuning methods. However, the privacy…

Incremental learning aims to overcome catastrophic forgetting when learning deep networks from sequential tasks. With impressive learning efficiency and performance, prompt-based methods adopt a fixed backbone to sequential tasks by…

The wide-ranging applications of large language models (LLMs), especially in safety-critical domains, necessitate the proper evaluation of the LLM's adversarial robustness. This paper proposes an efficient tool to audit the LLM's…

Backdoor attacks are an insidious security threat against machine learning models. Adversaries can manipulate the predictions of compromised models by inserting triggers into the training phase. Various backdoor attacks have been devised…

Chat template is a common technique used in the training and inference stages of Large Language Models (LLMs). It can transform input and output data into role-based and templated expressions to enhance the performance of LLMs. However,…

Graph Neural Networks(GNNs) are vulnerable to backdoor attacks, where adversaries implant malicious triggers to manipulate model predictions. Existing trigger generators are often simplistic in structure and overly reliant on specific…

Although multimodal large language models (MLLMs) are increasingly deployed in real-world applications, their instruction-following behavior leaves them vulnerable to prompt injection attacks. Existing prompt injection methods predominantly…

In this work, we demonstrate that certain machine unlearning methods may fail under straightforward prompt attacks. We systematically evaluate eight unlearning techniques across three model families using output-based, logit-based, and…

As the AI systems become deeply embedded in social media platforms, we've uncovered a concerning security vulnerability that goes beyond traditional adversarial attacks. It becomes important to assess the risks of LLMs before the general…

In this work, we explore "prompt tuning", a simple yet effective mechanism for learning "soft prompts" to condition frozen language models to perform specific downstream tasks. Unlike the discrete text prompts used by GPT-3, soft prompts…