Abstract:Attacking fairness is crucial because compromised models can introduce biased outcomes, undermining trust and amplifying inequalities in sensitive applications like hiring, healthcare, and law enforcement. This highlights the urgent need to understand how fairness mechanisms can be exploited and to develop defenses that ensure both fairness and robustness. We introduce BadFair, a novel backdoored fairness attack methodology. BadFair stealthily crafts a model that operates with accuracy and fairness under regular conditions but, when activated by certain triggers, discriminates and produces incorrect results for specific groups. This type of attack is particularly stealthy and dangerous, as it circumvents existing fairness detection methods, maintaining an appearance of fairness in normal use. Our findings reveal that BadFair achieves a more than 85% attack success rate in attacks aimed at target groups on average while only incurring a minimal accuracy loss. Moreover, it consistently exhibits a significant discrimination score, distinguishing between pre-defined target and non-target attacked groups across various datasets and models.
Abstract:This study identifies the potential vulnerabilities of Large Language Models (LLMs) to 'jailbreak' attacks, specifically focusing on the Arabic language and its various forms. While most research has concentrated on English-based prompt manipulation, our investigation broadens the scope to investigate the Arabic language. We initially tested the AdvBench benchmark in Standardized Arabic, finding that even with prompt manipulation techniques like prefix injection, it was insufficient to provoke LLMs into generating unsafe content. However, when using Arabic transliteration and chatspeak (or arabizi), we found that unsafe content could be produced on platforms like OpenAI GPT-4 and Anthropic Claude 3 Sonnet. Our findings suggest that using Arabic and its various forms could expose information that might remain hidden, potentially increasing the risk of jailbreak attacks. We hypothesize that this exposure could be due to the model's learned connection to specific words, highlighting the need for more comprehensive safety training across all language forms.
Abstract:It is imperative to ensure the stability of every prediction made by a language model; that is, a language's prediction should remain consistent despite minor input variations, like word substitutions. In this paper, we investigate the problem of certifying a language model's robustness against Universal Text Perturbations (UTPs), which have been widely used in universal adversarial attacks and backdoor attacks. Existing certified robustness based on random smoothing has shown considerable promise in certifying the input-specific text perturbations (ISTPs), operating under the assumption that any random alteration of a sample's clean or adversarial words would negate the impact of sample-wise perturbations. However, with UTPs, masking only the adversarial words can eliminate the attack. A naive method is to simply increase the masking ratio and the likelihood of masking attack tokens, but it leads to a significant reduction in both certified accuracy and the certified radius due to input corruption by extensive masking. To solve this challenge, we introduce a novel approach, the superior prompt search method, designed to identify a superior prompt that maintains higher certified accuracy under extensive masking. Additionally, we theoretically motivate why ensembles are a particularly suitable choice as base prompts for random smoothing. The method is denoted by superior prompt ensembling technique. We also empirically confirm this technique, obtaining state-of-the-art results in multiple settings. These methodologies, for the first time, enable high certified accuracy against both UTPs and ISTPs. The source code of CR-UTP is available at https://github.com/UCFML-Research/CR-UTP.
Abstract:Large Language Models (LLMs) are constrained by outdated information and a tendency to generate incorrect data, commonly referred to as "hallucinations." Retrieval-Augmented Generation (RAG) addresses these limitations by combining the strengths of retrieval-based methods and generative models. This approach involves retrieving relevant information from a large, up-to-date dataset and using it to enhance the generation process, leading to more accurate and contextually appropriate responses. Despite its benefits, RAG introduces a new attack surface for LLMs, particularly because RAG databases are often sourced from public data, such as the web. In this paper, we propose \TrojRAG{} to identify the vulnerabilities and attacks on retrieval parts (RAG database) and their indirect attacks on generative parts (LLMs). Specifically, we identify that poisoning several customized content passages could achieve a retrieval backdoor, where the retrieval works well for clean queries but always returns customized poisoned adversarial queries. Triggers and poisoned passages can be highly customized to implement various attacks. For example, a trigger could be a semantic group like "The Republican Party, Donald Trump, etc." Adversarial passages can be tailored to different contents, not only linked to the triggers but also used to indirectly attack generative LLMs without modifying them. These attacks can include denial-of-service attacks on RAG and semantic steering attacks on LLM generations conditioned by the triggers. Our experiments demonstrate that by just poisoning 10 adversarial passages can induce 98.2\% success rate to retrieve the adversarial passages. Then, these passages can increase the reject ratio of RAG-based GPT-4 from 0.01\% to 74.6\% or increase the rate of negative responses from 0.22\% to 72\% for targeted queries.
Abstract:Prompt tuning is one of the most effective solutions to adapting a fixed pre-trained language model (PLM) for various downstream tasks, especially with only a few input samples. However, the security issues, e.g., Trojan attacks, of prompt tuning on a few data samples are not well-studied. Transferring established data poisoning attacks directly to few-shot prompt tuning presents multiple challenges. One significant issue is the \textit{poisoned imbalance issue}, where non-target class samples are added to the target class, resulting in a greater number of target-class samples compared to non-target class. While this issue is not critical in regular tuning, it significantly hampers the few-shot prompt tuning, making it difficult to simultaneously achieve a high attack success rate (ASR) and maintain clean data accuracy (CDA). Additionally, few-shot prompting is prone to overfitting in terms of both ASR and CDA. In this paper, we introduce \textit{TrojFSP}, a method designed to address the challenges. To solve the poisoned imbalance issue, we develop a \textit{Target-Class Shrink (TC-Shrink)} technique, which aims to equalize the number of poisoning samples. To combat overfitting, we employ a \textit{Selective Token Poisoning} technique to boost attack performance. Furthermore, we introduce a \textit{Trojan-Trigger Attention} objective function to amplify the attention of the poisoned trojan prompt on triggers. Experiments show that our TrojFSP achieves an ASR of over 99\% while maintaining negligible decreases in CDA across various PLMs and datasets.
Abstract:Deep learning models have been incorporated into high-stakes sectors, including healthcare diagnosis, loan approvals, and candidate recruitment, among others. Consequently, any bias or unfairness in these models can harm those who depend on such models. In response, many algorithms have emerged to ensure fairness in deep learning. However, while the potential for harm is substantial, the resilience of these fair deep learning models against malicious attacks has never been thoroughly explored, especially in the context of emerging Trojan attacks. Moving beyond prior research, we aim to fill this void by introducing \textit{TrojFair}, a Trojan fairness attack. Unlike existing attacks, TrojFair is model-agnostic and crafts a Trojaned model that functions accurately and equitably for clean inputs. However, it displays discriminatory behaviors \text{-} producing both incorrect and unfair results \text{-} for specific groups with tainted inputs containing a trigger. TrojFair is a stealthy Fairness attack that is resilient to existing model fairness audition detectors since the model for clean inputs is fair. TrojFair achieves a target group attack success rate exceeding $88.77\%$, with an average accuracy loss less than $0.44\%$. It also maintains a high discriminative score between the target and non-target groups across various datasets and models.
Abstract:Prompt learning has been proven to be highly effective in improving pre-trained language model (PLM) adaptability, surpassing conventional fine-tuning paradigms, and showing exceptional promise in an ever-growing landscape of applications and APIs tailored for few-shot learning scenarios. Despite the growing prominence of prompt learning-based APIs, their security concerns remain underexplored. In this paper, we undertake a pioneering study on the Trojan susceptibility of prompt-learning PLM APIs. We identified several key challenges, including discrete-prompt, few-shot, and black-box settings, which limit the applicability of existing backdoor attacks. To address these challenges, we propose TrojPrompt, an automatic and black-box framework to effectively generate universal and stealthy triggers and insert Trojans into hard prompts. Specifically, we propose a universal API-driven trigger discovery algorithm for generating universal triggers for various inputs by querying victim PLM APIs using few-shot data samples. Furthermore, we introduce a novel progressive trojan poisoning algorithm designed to generate poisoned prompts that retain efficacy and transferability across a diverse range of models. Our experiments and results demonstrate TrojPrompt's capacity to effectively insert Trojans into text prompts in real-world black-box PLM APIs, while maintaining exceptional performance on clean test sets and significantly outperforming baseline models. Our work sheds light on the potential security risks in current models and offers a potential defensive approach.
Abstract:Self-supervised learning (SSL) is a commonly used approach to learning and encoding data representations. By using a pre-trained SSL image encoder and training a downstream classifier on top of it, impressive performance can be achieved on various tasks with very little labeled data. The increasing usage of SSL has led to an uptick in security research related to SSL encoders and the development of various Trojan attacks. The danger posed by Trojan attacks inserted in SSL encoders lies in their ability to operate covertly and spread widely among various users and devices. The presence of backdoor behavior in Trojaned encoders can inadvertently be inherited by downstream classifiers, making it even more difficult to detect and mitigate the threat. Although current Trojan detection methods in supervised learning can potentially safeguard SSL downstream classifiers, identifying and addressing triggers in the SSL encoder before its widespread dissemination is a challenging task. This is because downstream tasks are not always known, dataset labels are not available, and even the original training dataset is not accessible during the SSL encoder Trojan detection. This paper presents an innovative technique called SSL-Cleanse that is designed to detect and mitigate backdoor attacks in SSL encoders. We evaluated SSL-Cleanse on various datasets using 300 models, achieving an average detection success rate of 83.7% on ImageNet-100. After mitigating backdoors, on average, backdoored encoders achieve 0.24% attack success rate without great accuracy loss, proving the effectiveness of SSL-Cleanse.
Abstract:Vision Transformers (ViTs) have demonstrated the state-of-the-art performance in various vision-related tasks. The success of ViTs motivates adversaries to perform backdoor attacks on ViTs. Although the vulnerability of traditional CNNs to backdoor attacks is well-known, backdoor attacks on ViTs are seldom-studied. Compared to CNNs capturing pixel-wise local features by convolutions, ViTs extract global context information through patches and attentions. Na\"ively transplanting CNN-specific backdoor attacks to ViTs yields only a low clean data accuracy and a low attack success rate. In this paper, we propose a stealth and practical ViT-specific backdoor attack $TrojViT$. Rather than an area-wise trigger used by CNN-specific backdoor attacks, TrojViT generates a patch-wise trigger designed to build a Trojan composed of some vulnerable bits on the parameters of a ViT stored in DRAM memory through patch salience ranking and attention-target loss. TrojViT further uses minimum-tuned parameter update to reduce the bit number of the Trojan. Once the attacker inserts the Trojan into the ViT model by flipping the vulnerable bits, the ViT model still produces normal inference accuracy with benign inputs. But when the attacker embeds a trigger into an input, the ViT model is forced to classify the input to a predefined target class. We show that flipping only few vulnerable bits identified by TrojViT on a ViT model using the well-known RowHammer can transform the model into a backdoored one. We perform extensive experiments of multiple datasets on various ViT models. TrojViT can classify $99.64\%$ of test images to a target class by flipping $345$ bits on a ViT for ImageNet.