Related papers: Secret Stealing Attacks on Local LLM Fine-Tuning t…

Detecting semantic backdoors in classification models--where some classes can be activated by certain natural, but out-of-distribution inputs--is an important problem that has received relatively little attention. Semantic backdoors are…

Large language model fine-tuning APIs enable widespread model customization, yet pose significant safety risks. Recent work shows that adversaries can exploit access to these APIs to bypass model safety mechanisms by encoding harmful…

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…

Privacy-sensitive users require deploying large language models (LLMs) within their own infrastructure (on-premises) to safeguard private data and enable customization. However, vulnerabilities in local environments can lead to unauthorized…

Neural networks are widely known to be vulnerable to backdoor attacks, a method that poisons a portion of the training data to make the target model perform well on normal data sets, while outputting attacker-specified or random categories…

Aligned LLMs are secure, capable of recognizing and refusing to answer malicious questions. However, the role of internal parameters in maintaining such security is not well understood yet, further these models can be vulnerable to security…

The deployment of large language models (LLMs) on third-party devices requires new ways to protect model intellectual property. While Trusted Execution Environments (TEEs) offer a promising solution, their performance limits can lead to a…

Backdoor attacks pose a critical threat by embedding hidden triggers into inputs, causing models to misclassify them into target labels. While extensive research has focused on mitigating these attacks in object recognition models through…

By injecting a small number of poisoned samples into the training set, backdoor attacks aim to make the victim model produce designed outputs on any input injected with pre-designed backdoors. In order to achieve a high attack success rate…

The proliferation of open-weight Large Language Models (LLMs) has democratized agentic AI, yet fine-tuned weights are frequently shared and adopted with limited scrutiny beyond leaderboard performance. This creates a risk where third-party…

We reveal the theoretical foundations of techniques for editing large language models, and present new methods which can do so without requiring retraining. Our theoretical insights show that a single metric (a measure of the intrinsic…

Code generation large language models (LLMs) are increasingly integrated into modern software development workflows. Recent work has shown that these models are vulnerable to backdoor and poisoning attacks that induce the generation of…

The increasing use of large language models (LLMs) trained by third parties raises significant security concerns. In particular, malicious actors can introduce backdoors through poisoning attacks to generate undesirable outputs. While such…

With the boom in the natural language processing (NLP) field these years, backdoor attacks pose immense threats against deep neural network models. However, previous works hardly consider the effect of the poisoning rate. In this paper, our…

Large Language Models (LLMs) have transformed code completion tasks, providing context-based suggestions to boost developer productivity in software engineering. As users often fine-tune these models for specific applications, poisoning and…

Poisoning backdoor attacks involve an adversary manipulating the training data to induce certain behaviors in the victim model by inserting a trigger in the signal at inference time. We adapted clean label backdoor (CLBD)-data poisoning…

High-performance Deep Neural Networks (DNNs) are increasingly deployed in many real-world applications e.g., cloud prediction APIs. Recent advances in model functionality stealing attacks via black-box access (i.e., inputs in, predictions…

Detecting whether a model has been poisoned is a longstanding problem in AI security. In this work, we present a practical scanner for identifying sleeper agent-style backdoors in causal language models. Our approach relies on two key…

We introduce a new class of attacks on machine learning models. We show that an adversary who can poison a training dataset can cause models trained on this dataset to leak significant private details of training points belonging to other…

Backdoor attacks pose a significant threat to deep neural networks, particularly as recent advancements have led to increasingly subtle implantation, making the defense more challenging. Existing defense mechanisms typically rely on an…