Related papers: Strong Backdoors for Default Logic
Backdoor attacks are an insidious security threat against machine learning models. Adversaries can manipulate the predictions of compromised models by inserting triggers into the training phase. Various backdoor attacks have been devised…
The increasing importance of both deep neural networks (DNNs) and cloud services for training them means that bad actors have more incentive and opportunity to insert backdoors to alter the behavior of trained models. In this paper, we…
The success of deep neural networks (DNNs) in real-world applications has benefited from abundant pre-trained models. However, the backdoored pre-trained models can pose a significant trojan threat to the deployment of downstream DNNs.…
Backdoors and backbones of Boolean formulas are hidden structural properties. A natural goal, already in part realized, is that solver algorithms seek to obtain substantially better performance by exploiting these structures. However, the…
Knuth (1990) introduced the class of nested formulas and showed that their satisfiability can be decided in polynomial time. We show that, parameterized by the size of a smallest strong backdoor set to the target class of nested formulas,…
Intuitively, a backdoor attack against Deep Neural Networks (DNNs) is to inject hidden malicious behaviors into DNNs such that the backdoor model behaves legitimately for benign inputs, yet invokes a predefined malicious behavior when its…
A backdoor deep learning (DL) model behaves normally upon clean inputs but misbehaves upon trigger inputs as the backdoor attacker desires, posing severe consequences to DL model deployments. State-of-the-art defenses are either limited to…
Despite the notable advancements of existing prompting methods, such as In-Context Learning and Chain-of-Thought for Large Language Models (LLMs), they still face challenges related to various biases. Traditional debiasing methods primarily…
We present Sparse Backdoor, a supply-chain attack that plants a \emph{provably undetectable} backdoor in pre-trained image classifiers, including convolutional networks and Vision Transformers. The attack injects a structured sparse…
Deep Neural Networks (DNNs) are susceptible to backdoor attacks during training. The model corrupted in this way functions normally, but when triggered by certain patterns in the input, produces a predefined target label. Existing defenses…
The introduction of robust optimisation has pushed the state-of-the-art in defending against adversarial attacks. Notably, the state-of-the-art projected gradient descent (PGD)-based training method has been shown to be universally and…
This paper proposes a data-efficient detection method for deep neural networks against backdoor attacks under a black-box scenario. The proposed approach is motivated by the intuition that features corresponding to triggers have a higher…
The programming language Prolog makes declarative programming possible, at least to a substantial extent. Programs may be written and reasoned about in terms of their declarative semantics. All the advantages of declarative programming are…
Data-poisoning backdoor attacks are serious security threats to machine learning models, where an adversary can manipulate the training dataset to inject backdoors into models. In this paper, we focus on in-training backdoor defense, aiming…
Backdoor learning is a critical research topic for understanding the vulnerabilities of deep neural networks. While the diffusion model (DM) has been broadly deployed in public over the past few years, the understanding of its backdoor…
Backdoor attacks are a significant threat to the performance and integrity of pre-trained language models. Although such models are routinely fine-tuned for downstream NLP tasks, recent work shows they remain vulnerable to backdoor attacks…
Although backdoor learning is an active research topic in the NLP domain, the literature lacks studies that systematically categorize and summarize backdoor attacks and defenses. To bridge the gap, we present a comprehensive and unifying…
Backdoor attacks pose a significant security threat to natural language processing (NLP) systems, but existing methods lack explainable trigger mechanisms and fail to quantitatively model vulnerability patterns. This work pioneers the…
The deployment of Large Reasoning Models (LRMs) in high-stakes decision-making pipelines has introduced a novel and opaque attack surface: reasoning backdoors. In these attacks, the model's intermediate Chain-of-Thought (CoT) is manipulated…
Abductive reasoning (or Abduction, for short) is among the most fundamental AI reasoning methods, with a broad range of applications, including fault diagnosis, belief revision, and automated planning. Unfortunately, Abduction is of high…