BlueSuffix: Reinforced Blue Teaming for Vision-Language Models Against Jailbreak Attacks

Despite their superb multimodal capabilities, Vision-Language Models (VLMs) have been shown to be vulnerable to jailbreak attacks, which are inference-time attacks that induce the model to output harmful responses with tricky prompts. It is thus essential to defend VLMs against potential jailbreaks for their trustworthy deployment in real-world applications. In this work, we focus on black-box defense for VLMs against jailbreak attacks. Existing black-box defense methods are either unimodal or bimodal. Unimodal methods enhance either the vision or language module of the VLM, while bimodal methods robustify the model through text-image representation realignment. However, these methods suffer from two limitations: 1) they fail to fully exploit the cross-modal information, or 2) they degrade the model performance on benign inputs. To address these limitations, we propose a novel blue-team method BlueSuffix that defends the black-box target VLM against jailbreak attacks without compromising its performance. BlueSuffix includes three key components: 1) a visual purifier against jailbreak images, 2) a textual purifier against jailbreak texts, and 3) a blue-team suffix generator fine-tuned via reinforcement learning for enhancing cross-modal robustness. We empirically show on three VLMs (LLaVA, MiniGPT-4, and Gemini) and two safety benchmarks (MM-SafetyBench and RedTeam-2K) that BlueSuffix outperforms the baseline defenses by a significant margin. Our BlueSuffix opens up a promising direction for defending VLMs against jailbreak attacks.

Expose Before You Defend: Unifying and Enhancing Backdoor Defenses via Exposed Models

Backdoor attacks covertly implant triggers into deep neural networks (DNNs) by poisoning a small portion of the training data with pre-designed backdoor triggers. This vulnerability is exacerbated in the era of large models, where extensive (pre-)training on web-crawled datasets is susceptible to compromise. In this paper, we introduce a novel two-step defense framework named Expose Before You Defend (EBYD). EBYD unifies existing backdoor defense methods into a comprehensive defense system with enhanced performance. Specifically, EBYD first exposes the backdoor functionality in the backdoored model through a model preprocessing step called backdoor exposure, and then applies detection and removal methods to the exposed model to identify and eliminate the backdoor features. In the first step of backdoor exposure, we propose a novel technique called Clean Unlearning (CUL), which proactively unlearns clean features from the backdoored model to reveal the hidden backdoor features. We also explore various model editing/modification techniques for backdoor exposure, including fine-tuning, model sparsification, and weight perturbation. Using EBYD, we conduct extensive experiments on 10 image attacks and 6 text attacks across 2 vision datasets (CIFAR-10 and an ImageNet subset) and 4 language datasets (SST-2, IMDB, Twitter, and AG's News). The results demonstrate the importance of backdoor exposure for backdoor defense, showing that the exposed models can significantly benefit a range of downstream defense tasks, including backdoor label detection, backdoor trigger recovery, backdoor model detection, and backdoor removal. We hope our work could inspire more research in developing advanced defense frameworks with exposed models. Our code is available at: https://github.com/bboylyg/Expose-Before-You-Defend.

AnyAttack: Towards Large-scale Self-supervised Generation of Targeted Adversarial Examples for Vision-Language Models

Due to their multimodal capabilities, Vision-Language Models (VLMs) have found numerous impactful applications in real-world scenarios. However, recent studies have revealed that VLMs are vulnerable to image-based adversarial attacks, particularly targeted adversarial images that manipulate the model to generate harmful content specified by the adversary. Current attack methods rely on predefined target labels to create targeted adversarial attacks, which limits their scalability and applicability for large-scale robustness evaluations. In this paper, we propose AnyAttack, a self-supervised framework that generates targeted adversarial images for VLMs without label supervision, allowing any image to serve as a target for the attack. To address the limitation of existing methods that require label supervision, we introduce a contrastive loss that trains a generator on a large-scale unlabeled image dataset, LAION-400M dataset, for generating targeted adversarial noise. This large-scale pre-training endows our method with powerful transferability across a wide range of VLMs. Extensive experiments on five mainstream open-source VLMs (CLIP, BLIP, BLIP2, InstructBLIP, and MiniGPT-4) across three multimodal tasks (image-text retrieval, multimodal classification, and image captioning) demonstrate the effectiveness of our attack. Additionally, we successfully transfer AnyAttack to multiple commercial VLMs, including Google's Gemini, Claude's Sonnet, and Microsoft's Copilot. These results reveal an unprecedented risk to VLMs, highlighting the need for effective countermeasures.

Adversarial Suffixes May Be Features Too!

Despite significant ongoing efforts in safety alignment, large language models (LLMs) such as GPT-4 and LLaMA 3 remain vulnerable to jailbreak attacks that can induce harmful behaviors, including those triggered by adversarial suffixes. Building on prior research, we hypothesize that these adversarial suffixes are not mere bugs but may represent features that can dominate the LLM's behavior. To evaluate this hypothesis, we conduct several experiments. First, we demonstrate that benign features can be effectively made to function as adversarial suffixes, i.e., we develop a feature extraction method to extract sample-agnostic features from benign dataset in the form of suffixes and show that these suffixes may effectively compromise safety alignment. Second, we show that adversarial suffixes generated from jailbreak attacks may contain meaningful features, i.e., appending the same suffix to different prompts results in responses exhibiting specific characteristics. Third, we show that such benign-yet-safety-compromising features can be easily introduced through fine-tuning using only benign datasets, i.e., even in the absence of harmful content. This highlights the critical risk posed by dominating benign features in the training data and calls for further research to reinforce LLM safety alignment. Our code and data is available at \url{https://github.com/anonymous}.

Do Influence Functions Work on Large Language Models?

Influence functions aim to quantify the impact of individual training data points on a model's predictions. While extensive research has been conducted on influence functions in traditional machine learning models, their application to large language models (LLMs) has been limited. In this work, we conduct a systematic study to address a key question: do influence functions work on LLMs? Specifically, we evaluate influence functions across multiple tasks and find that they consistently perform poorly in most settings. Our further investigation reveals that their poor performance can be attributed to: (1) inevitable approximation errors when estimating the iHVP component due to the scale of LLMs, (2) uncertain convergence during fine-tuning, and, more fundamentally, (3) the definition itself, as changes in model parameters do not necessarily correlate with changes in LLM behavior. Our study thus suggests the need for alternative approaches for identifying influential samples. To support future work, our code is made available at https://github.com/plumprc/Failures-of-Influence-Functions-in-LLMs.

BackdoorLLM: A Comprehensive Benchmark for Backdoor Attacks on Large Language Models

Generative Large Language Models (LLMs) have made significant strides across various tasks, but they remain vulnerable to backdoor attacks, where specific triggers in the prompt cause the LLM to generate adversary-desired responses. While most backdoor research has focused on vision or text classification tasks, backdoor attacks in text generation have been largely overlooked. In this work, we introduce \textit{BackdoorLLM}, the first comprehensive benchmark for studying backdoor attacks on LLMs. \textit{BackdoorLLM} features: 1) a repository of backdoor benchmarks with a standardized training pipeline, 2) diverse attack strategies, including data poisoning, weight poisoning, hidden state attacks, and chain-of-thought attacks, 3) extensive evaluations with over 200 experiments on 8 attacks across 7 scenarios and 6 model architectures, and 4) key insights into the effectiveness and limitations of backdoors in LLMs. We hope \textit{BackdoorLLM} will raise awareness of backdoor threats and contribute to advancing AI safety. The code is available at \url{https://github.com/bboylyg/BackdoorLLM}.

Multi-Trigger Backdoor Attacks: More Triggers, More Threats

Backdoor attacks have emerged as a primary threat to (pre-)training and deployment of deep neural networks (DNNs). While backdoor attacks have been extensively studied in a body of works, most of them were focused on single-trigger attacks that poison a dataset using a single type of trigger. Arguably, real-world backdoor attacks can be much more complex, e.g., the existence of multiple adversaries for the same dataset if it is of high value. In this work, we investigate the practical threat of backdoor attacks under the setting of \textbf{multi-trigger attacks} where multiple adversaries leverage different types of triggers to poison the same dataset. By proposing and investigating three types of multi-trigger attacks, including parallel, sequential, and hybrid attacks, we provide a set of important understandings of the coexisting, overwriting, and cross-activating effects between different triggers on the same dataset. Moreover, we show that single-trigger attacks tend to cause overly optimistic views of the security of current defense techniques, as all examined defense methods struggle to defend against multi-trigger attacks. Finally, we create a multi-trigger backdoor poisoning dataset to help future evaluation of backdoor attacks and defenses. Although our work is purely empirical, we hope it can help steer backdoor research toward more realistic settings.

End-to-End Anti-Backdoor Learning on Images and Time Series

Backdoor attacks present a substantial security concern for deep learning models, especially those utilized in applications critical to safety and security. These attacks manipulate model behavior by embedding a hidden trigger during the training phase, allowing unauthorized control over the model's output during inference time. Although numerous defenses exist for image classification models, there is a conspicuous absence of defenses tailored for time series data, as well as an end-to-end solution capable of training clean models on poisoned data. To address this gap, this paper builds upon Anti-Backdoor Learning (ABL) and introduces an innovative method, End-to-End Anti-Backdoor Learning (E2ABL), for robust training against backdoor attacks. Unlike the original ABL, which employs a two-stage training procedure, E2ABL accomplishes end-to-end training through an additional classification head linked to the shallow layers of a Deep Neural Network (DNN). This secondary head actively identifies potential backdoor triggers, allowing the model to dynamically cleanse these samples and their corresponding labels during training. Our experiments reveal that E2ABL significantly improves on existing defenses and is effective against a broad range of backdoor attacks in both image and time series domains.

Reconstructive Neuron Pruning for Backdoor Defense

Deep neural networks (DNNs) have been found to be vulnerable to backdoor attacks, raising security concerns about their deployment in mission-critical applications. While existing defense methods have demonstrated promising results, it is still not clear how to effectively remove backdoor-associated neurons in backdoored DNNs. In this paper, we propose a novel defense called \emph{Reconstructive Neuron Pruning} (RNP) to expose and prune backdoor neurons via an unlearning and then recovering process. Specifically, RNP first unlearns the neurons by maximizing the model's error on a small subset of clean samples and then recovers the neurons by minimizing the model's error on the same data. In RNP, unlearning is operated at the neuron level while recovering is operated at the filter level, forming an asymmetric reconstructive learning procedure. We show that such an asymmetric process on only a few clean samples can effectively expose and prune the backdoor neurons implanted by a wide range of attacks, achieving a new state-of-the-art defense performance. Moreover, the unlearned model at the intermediate step of our RNP can be directly used to improve other backdoor defense tasks including backdoor removal, trigger recovery, backdoor label detection, and backdoor sample detection. Code is available at \url{https://github.com/bboylyg/RNP}.

Anti-Backdoor Learning: Training Clean Models on Poisoned Data

Backdoor attack has emerged as a major security threat to deep neural networks (DNNs). While existing defense methods have demonstrated promising results on detecting or erasing backdoors, it is still not clear whether robust training methods can be devised to prevent the backdoor triggers being injected into the trained model in the first place. In this paper, we introduce the concept of \emph{anti-backdoor learning}, aiming to train \emph{clean} models given backdoor-poisoned data. We frame the overall learning process as a dual-task of learning the \emph{clean} and the \emph{backdoor} portions of data. From this view, we identify two inherent characteristics of backdoor attacks as their weaknesses: 1) the models learn backdoored data much faster than learning with clean data, and the stronger the attack the faster the model converges on backdoored data; 2) the backdoor task is tied to a specific class (the backdoor target class). Based on these two weaknesses, we propose a general learning scheme, Anti-Backdoor Learning (ABL), to automatically prevent backdoor attacks during training. ABL introduces a two-stage \emph{gradient ascent} mechanism for standard training to 1) help isolate backdoor examples at an early training stage, and 2) break the correlation between backdoor examples and the target class at a later training stage. Through extensive experiments on multiple benchmark datasets against 10 state-of-the-art attacks, we empirically show that ABL-trained models on backdoor-poisoned data achieve the same performance as they were trained on purely clean data. Code is available at \url{https://github.com/bboylyg/ABL}.