45 accepted papers
The tendency to find and exploit "shortcuts" to complete tasks poses significant risks for reliable assessment and deployment of large language models (LLMs). For example, an LLM agent with access to unit tests may delete failing tests rather than fix the underlying bug. Such behavior undermines bot…
In the past decade, considerable research effort has been devoted to securing machine learning (ML) models that operate in adversarial settings. Yet, progress has been slow even for simple "toy" problems (e.g., robustness to small adversarial perturbations) and is often hindered by non-rigorous eval…
AI agents may soon become capable of autonomously completing valuable, long-horizon tasks in diverse domains. Current benchmarks either do not measure real-world tasks, or are not sufficiently difficult to meaningfully measure frontier models. To this end, we present Terminal-Bench 1.5: a carefully…
Artists are increasingly concerned about advancements in image generation models that can closely replicate their unique artistic styles. In response, several protection tools against style mimicry have been developed that incorporate small adversarial perturbations into artworks published online. I…
We introduce AutoAdvExBench, a benchmark to evaluate if large language models (LLMs) can autonomously exploit defenses to adversarial examples. Unlike existing security benchmarks that often serve as proxies for real-world tasks, AutoAdvExBench directly measures LLMs' success on tasks regularly perf…
It is now common to evaluate Large Language Models (LLMs) by having humans manually vote to evaluate model outputs, in contrast to typical benchmarks that evaluate knowledge or skill at some particular task. Chatbot Arena, the most popular benchmark of this type, ranks models by asking users to sele…
We study how training data contributes to the emergence of toxic behaviors in large language models. Most prior work on reducing model toxicity adopts *reactive* approaches, such as fine-tuning pre-trained (and potentially toxic) models to align them with human values. In contrast, we propose a *pro…
Large language models memorize parts of their training data. Memorizing short snippets and facts is required to answer questions about the world and to be fluent in any language. But models have also been shown to reproduce long verbatim sequences of memorized text when prompted by a motivated adver…
Many stakeholders---from model developers to policymakers---seek to minimize the risks of large language models (LLMs). Key to this goal is whether technical safeguards can impede the misuse of LLMs, even when models are customizable via fine-tuning or when model weights are openly available. Severa…
Large language models are pre-trained on uncurated text datasets consisting of trillions of tokens scraped from the Web. Prior work has shown that: (1) web-scraped pre-training datasets can be practically poisoned by malicious actors; and (2) adversaries can compromise language models after poisonin…
The widespread deployment of general-purpose AI (GPAI) systems introduces significant new risks. Yet the infrastructure, practices, and norms for reporting flaws in GPAI systems remain seriously underdeveloped, lagging far behind more established fields like software security. Based on a collaborati…
Large language models are prone to *memorizing* some of their training data. Memorized (and possibly sensitive) samples can then be extracted at generation time by adversarial or benign users. There is hope that *model alignment*---a standard training process that tunes a model to harmlessly follow…
With the prevalence of the Pretraining-Finetuning paradigm in transfer learning the robustness of downstream tasks has become a critical concern. In this work we delve into adversarial robustness in transfer learning and reveal the critical role of initialization including both the pretrained model…
The performance of differentially private machine learning can be boosted significantly by leveraging the transfer learning capabilities of non-private models pretrained on large *public* datasets. We critically review this approach. We primarily question whether the use of large Web-scraped dataset…
It is commonplace to produce application-specific models by fine-tuning large pre-trained models using a small bespoke dataset. The widespread availability of foundation model checkpoints on the web poses considerable risks, including the vulnerability to backdoor attacks. In this paper, we unveil a…
Recent work has shown it is possible to construct adversarial examples that cause aligned language models to emit harmful strings or perform harmful behavior. Existing attacks work either in the white-box setting (with full access to the model weights), or through _transferability_: the phenomenon t…
We introduce the first model-stealing attack that extracts precise, nontrivial information from black-box production language models like OpenAI's ChatGPT or Google's PaLM-2. Specifically, our attack recovers the embedding projection layer (up to symmetries) of a transformer model, given typical API…
In this paper we show how to achieve state-of-the-art certified adversarial robustness to 2-norm bounded perturbations by relying exclusively on off-the-shelf pretrained models. To do so, we instantiate the denoised smoothing approach of Salman et al. by combining a pretrained denoising diffusion pr…
Large language models are now tuned to align with the goals of their creators, namely to be "helpful and harmless." These models should respond helpfully to user questions, but refuse to answer requests that could cause harm. However, adversarial users can construct inputs which circumvent attempts…
Modern neural language models that are widely used in various NLP tasks risk memorizing sensitive information from their training data. Understanding this memorization is important in real world applications and also from a learning-theoretical perspective. An open question in previous studies of la…
``Effective robustness'' measures the extra out-of-distribution (OOD) robustness beyond what can be predicted from the in-distribution (ID) performance. Existing effective robustness evaluations typically use a single test set such as ImageNet to evaluate the ID accuracy. This becomes problematic wh…
Machine learning models exhibit two seemingly contradictory phenomena: training data memorization and various forms of forgetting. In memorization, models overfit specific training examples and become susceptible to privacy attacks. In forgetting, examples which appeared early in training are forgot…
We show that combining human prior knowledge with end-to-end learning can improve the robustness of deep neural networks by introducing a part-based model for object classification. We believe that the richer form of annotation helps guide neural networks to learn more robust features without requir…
Decision-based attacks construct adversarial examples against a machine learning (ML) model by making only hard-label queries. These attacks have mainly been applied directly to standalone neural networks. However, in practice, ML models are just one component of a larger learning system. We find th…
Large language models (LMs) have been shown to memorize parts of their training data, and when prompted appropriately, they will emit the memorized training data verbatim. This is undesirable because memorization violates privacy (exposing user data), degrades utility (repeated easy-to-memorize text…
Model distillation is frequently proposed as a technique to reduce the privacy leakage of machine learning. These empirical privacy defenses rely on the intuition that distilled ``student'' models protect the privacy of training data, as they only interact with this data indirectly through a ``teach…
We extend semi-supervised learning to the problem of domain adaptation to learn significantly higher-accuracy models that train on one data distribution and test on a different one. With the goal of generality, we introduce AdaMatch, a unified solution for unsupervised domain adaptation (UDA), semi-…
Data poisoning has been proposed as a compelling defense against facial recognition models trained on Web-scraped pictures. Users can perturb images they post online, so that models will misclassify future (unperturbed) pictures. We demonstrate that this strategy provides a false sense of securi…
We find that existing language modeling datasets contain many near-duplicate examples and long repetitive substrings. As a result, over 1% of the unprompted output of language models trained on these datasets is copied verbatim from the training data. We develop two tools that allow us to deduplicat…
Evading adversarial example detection defenses requires finding adversarial examples that must simultaneously (a) be misclassified by the model and (b) be detected as non-adversarial. We find that existing attacks that attempt to satisfy multiple simultaneous constraints often over-optimize against…
When machine learning training is outsourced to third parties, $backdoor$ $attacks$ become practical as the third party who trains the model may act maliciously to inject hidden behaviors into the otherwise accurate model. Until now, the mechanism to inject backdoors has been limited to $poisoning$.…
Hundreds of defenses have been proposed to make deep neural networks robust against minimal (adversarial) input perturbations. However, only a handful of these defenses held up their claims because correctly evaluating robustness is extremely challenging: Weak attacks often fail to find adversarial…
Evaluating robustness of machine-learning models to adversarial examples is a challenging problem. Many defenses have been shown to provide a false sense of robustness by causing gradient-based attacks to fail, and they have been broken under more rigorous evaluations. Although guidelines and best p…
Multimodal contrastive learning methods like CLIP train on noisy and uncurated training datasets. This is cheaper than labeling datasets manually, and even improves out-of-distribution robustness. We show that this practice makes backdoor and poisoning attacks a significant threat. By poisoning just…
Machine learning models trained on private datasets have been shown to leak their private data. Recent work has found that the average data point is rarely leaked---it is often the outlier samples that are subject to memorization and, consequently, leakage. We demonstrate and analyze an Onion Effect…
Membership inference is one of the simplest privacy threats faced by machine learning models that are trained on private sensitive data. In this attack, an adversary infers whether a particular point was used to train the model, or not, by observing the model’s predictions. Whereas current attack me…
Semi-supervised learning (SSL) provides an effective means of leveraging unlabeled data to improve a model’s performance. This domain has seen fast progress recently, at the cost of requiring more complex methods. In this paper we propose FixMatch, an algorithm that is a significant simplification o…
Adversarial examples are malicious inputs crafted to induce misclassification. Commonly studied \emph{sensitivity-based} adversarial examples introduce semantically-small changes to an input that result in a different model prediction. This paper studies a complementary failure mode, \emph{invarianc…
We study how robust current ImageNet models are to distribution shifts arising from natural variations in datasets. Most research on robustness focuses on synthetic image perturbations (noise, simulated weather artifacts, adversarial examples, etc.), which leaves open how robustness on synthetic dis…
Adaptive attacks have (rightfully) become the de facto standard for evaluating defenses to adversarial examples. We find, however, that typical adaptive evaluations are incomplete. We demonstrate that 13 defenses recently published at ICLR, ICML and NeurIPS---and which illustrate a diverse set of de…
We improve the recently-proposed ``MixMatch semi-supervised learning algorithm by introducing two new techniques: distribution alignment and augmentation anchoring. - Distribution alignment encourages the marginal distribution of predictions on unlabeled data to be close to the marginal distribution…
Over the last few years, the phenomenon of
Adversarial examples are inputs to machine learning models designed by an adversary to cause an incorrect output. So far, adversarial examples have been studied most extensively in the image domain. In this domain, adversarial examples can be constructed by imperceptibly modifying images to cause mi…
Semi-supervised learning has proven to be a powerful paradigm for leveraging unlabeled data to mitigate the reliance on large labeled datasets. In this work, we unify the current dominant approaches for semi-supervised learning to produce a new algorithm, MixMatch, that guesses low-entropy labels fo…
We identify obfuscated gradients, a kind of gradient masking, as a phenomenon that leads to a false sense of security in defenses against adversarial examples. While defenses that cause obfuscated gradients appear to defeat iterative optimization-based attacks, we find defenses relying on this effec…