Multi-modal encoders map images, sounds, texts, videos, etc. into a single embedding space, aligning representations across modalities (e.g., associate an image of a dog with a barking sound). We show that multi-modal embeddings can be vulnerable to an attack we call "adversarial illusions." Given an input in any modality, an adversary can perturb it so as to make its embedding close to that of an arbitrary, adversary-chosen input in another modality. Illusions thus enable the adversary to align any image with any text, any text with any sound, etc. Adversarial illusions exploit proximity in the embedding space and are thus agnostic to downstream tasks. Using ImageBind embeddings, we demonstrate how adversarially aligned inputs, generated without knowledge of specific downstream tasks, mislead image generation, text generation, and zero-shot classification.
We demonstrate how images and sounds can be used for indirect prompt and instruction injection in multi-modal LLMs. An attacker generates an adversarial perturbation corresponding to the prompt and blends it into an image or audio recording. When the user asks the (unmodified, benign) model about the perturbed image or audio, the perturbation steers the model to output the attacker-chosen text and/or make the subsequent dialog follow the attacker's instruction. We illustrate this attack with several proof-of-concept examples targeting LLaVa and PandaGPT.
Commoditization and broad adoption of machine learning (ML) technologies expose users of these technologies to new security risks. Many models today are based on neural networks. Training and deploying these models for real-world applications involves complex hardware and software pipelines applied to training data from many sources. Models trained on untrusted data are vulnerable to poisoning attacks that introduce "backdoor" functionality. Compromising a fraction of the training data requires few resources from the attacker, but defending against these attacks is a challenge. Although there have been dozens of defenses proposed in the research literature, most of them are expensive to integrate or incompatible with the existing training pipelines. In this paper, we take a pragmatic, developer-centric view and show how practitioners can answer two actionable questions: (1) how robust is my model to backdoor poisoning attacks?, and (2) how can I make it more robust without changing the training pipeline? We focus on the size of the compromised subset of the training data as a universal metric. We propose an easy-to-learn primitive sub-task to estimate this metric, thus providing a baseline on backdoor poisoning. Next, we show how to leverage hyperparameter search - a tool that ML developers already extensively use - to balance the model's accuracy and robustness to poisoning, without changes to the training pipeline. We demonstrate how to use our metric to estimate the robustness of models to backdoor attacks. We then design, implement, and evaluate a multi-stage hyperparameter search method we call Mithridates that strengthens robustness by 3-5x with only a slight impact on the model's accuracy. We show that the hyperparameters found by our method increase robustness against multiple types of backdoor attacks and extend our method to AutoML and federated learning.
Federated learning with differential privacy, i.e. private federated learning (PFL), makes it possible to train models on private data distributed across users' devices without harming privacy. PFL is efficient for models, such as neural networks, that have a fixed number of parameters, and thus a fixed-dimensional gradient vector. Such models include neural-net language models, but not tokenizers, the topic of this work. Training a tokenizer requires frequencies of words from an unlimited vocabulary, and existing methods for finding an unlimited vocabulary need a separate privacy budget. A workaround is to train the tokenizer on publicly available data. However, in this paper we first show that a tokenizer trained on mismatched data results in worse model performance compared to a privacy-violating "oracle" tokenizer that accesses user data, with perplexity increasing by 20%. We also show that sub-word tokenizers are better suited to the federated context than word-level ones, since they can encode new words, though with more tokens per word. Second, we propose a novel method to obtain a tokenizer without using any additional privacy budget. During private federated learning of the language model, we sample from the model, train a new tokenizer on the sampled sequences, and update the model embeddings. We then continue private federated learning, and obtain performance within 1% of the "oracle" tokenizer. Since this process trains the tokenizer only indirectly on private data, we can use the "postprocessing guarantee" of differential privacy and thus use no additional privacy budget.
We investigate a new threat to neural sequence-to-sequence (seq2seq) models: training-time attacks that cause models to "spin" their outputs so as to support an adversary-chosen sentiment or point of view, but only when the input contains adversary-chosen trigger words. For example, a spinned summarization model would output positive summaries of any text that mentions the name of some individual or organization. Model spinning enables propaganda-as-a-service. An adversary can create customized language models that produce desired spins for chosen triggers, then deploy them to generate disinformation (a platform attack), or else inject them into ML training pipelines (a supply-chain attack), transferring malicious functionality to downstream models. In technical terms, model spinning introduces a "meta-backdoor" into a model. Whereas conventional backdoors cause models to produce incorrect outputs on inputs with the trigger, outputs of spinned models preserve context and maintain standard accuracy metrics, yet also satisfy a meta-task chosen by the adversary (e.g., positive sentiment). To demonstrate feasibility of model spinning, we develop a new backdooring technique. It stacks the adversarial meta-task onto a seq2seq model, backpropagates the desired meta-task output to points in the word-embedding space we call "pseudo-words," and uses pseudo-words to shift the entire output distribution of the seq2seq model. We evaluate this attack on language generation, summarization, and translation models with different triggers and meta-tasks such as sentiment, toxicity, and entailment. Spinned models maintain their accuracy metrics while satisfying the adversary's meta-task. In supply chain attack the spin transfers to downstream models. Finally, we propose a black-box, meta-task-independent defense to detect models that selectively apply spin to inputs with a certain trigger.
We design a scalable algorithm to privately generate location heatmaps over decentralized data from millions of user devices. It aims to ensure differential privacy before data becomes visible to a service provider while maintaining high data accuracy and minimizing resource consumption on users' devices. To achieve this, we revisit the distributed differential privacy concept based on recent results in the secure multiparty computation field and design a scalable and adaptive distributed differential privacy approach for location analytics. Evaluation on public location datasets shows that this approach successfully generates metropolitan-scale heatmaps from millions of user samples with a worst-case client communication overhead that is significantly smaller than existing state-of-the-art private protocols of similar accuracy.
We investigate a new threat to neural sequence-to-sequence (seq2seq) models: training-time attacks that cause models to "spin" their output and support a certain sentiment when the input contains adversary-chosen trigger words. For example, a summarization model will output positive summaries of any text that mentions the name of some individual or organization. We introduce the concept of a "meta-backdoor" to explain model-spinning attacks. These attacks produce models whose output is valid and preserves context, yet also satisfies a meta-task chosen by the adversary (e.g., positive sentiment). Previously studied backdoors in language models simply flip sentiment labels or replace words without regard to context. Their outputs are incorrect on inputs with the trigger. Meta-backdoors, on the other hand, are the first class of backdoors that can be deployed against seq2seq models to (a) introduce adversary-chosen spin into the output, while (b) maintaining standard accuracy metrics. To demonstrate feasibility of model spinning, we develop a new backdooring technique. It stacks the adversarial meta-task (e.g., sentiment analysis) onto a seq2seq model, backpropagates the desired meta-task output (e.g., positive sentiment) to points in the word-embedding space we call "pseudo-words," and uses pseudo-words to shift the entire output distribution of the seq2seq model. Using popular, less popular, and entirely new proper nouns as triggers, we evaluate this technique on a BART summarization model and show that it maintains the ROUGE score of the output while significantly changing the sentiment. We explain why model spinning can be a dangerous technique in AI-powered disinformation and discuss how to mitigate these attacks.
We investigate a new method for injecting backdoors into machine learning models, based on poisoning the loss computation in the model-training code. Our attack is blind: the attacker cannot modify the training data, nor observe the execution of his code, nor access the resulting model. We develop a new technique for blind backdoor training using multi-objective optimization to achieve high accuracy on both the main and backdoor tasks while evading all known defenses. We then demonstrate the efficacy of the blind attack with new classes of backdoors strictly more powerful than those in prior literature: single-pixel backdoors in ImageNet models, backdoors that switch the model to a different, complex task, and backdoors that do not require inference-time input modifications. Finally, we discuss defenses.
We are increasingly surrounded by applications, connected devices, services, and smart environments which require fine-grained access to various personal data. The inherent complexities of our personal and professional policies and preferences in interactions with these analytics services raise important challenges in privacy. Moreover, due to sensitivity of the data and regulatory and technical barriers, it is not always feasible to do these policy negotiations in a centralized manner. In this paper we present PoliBox, a decentralized, edge-based framework for policy-based personal data analytics. PoliBox brings together a number of existing established components to provide privacy-preserving analytics within a distributed setting. We evaluate our framework using a popular exemplar of private analytics, Federated Learning, and demonstrate that for varying model sizes and use cases, PoliBox is able to perform accurate model training and inference within very reasonable resource and time budgets.
Federated learning (FL) is a heavily promoted approach for training ML models on sensitive data, e.g., text typed by users on their smartphones. FL is expressly designed for training on data that are unbalanced and non-iid across the participants. To ensure privacy and integrity of the federated model, latest FL approaches use differential privacy or robust aggregation to limit the influence of "outlier" participants. First, we show that on standard tasks such as next-word prediction, many participants gain no benefit from FL because the federated model is less accurate on their data than the models they can train locally on their own. Second, we show that differential privacy and robust aggregation make this problem worse by further destroying the accuracy of the federated model for many participants. Then, we evaluate three techniques for local adaptation of federated models: fine-tuning, multi-task learning, and knowledge distillation. We analyze where each technique is applicable and demonstrate that all participants benefit from local adaptation. Participants whose local models are poor obtain big accuracy improvements over conventional FL. Participants whose local models are better than the federated model and who have no incentive to participate in FL today improve less, but sufficiently to make the adapted federated model better than their local models.