Related papers: Fast algebraic immunity of Boolean functions and L…
Function-correcting codes are an innovative class of codes that are designed to protect a function evaluation of the data against errors or corruptions. Due to its usefulness in machine learning applications and archival data storage, where…
Although safely enhanced Large Language Models (LLMs) have achieved remarkable success in tackling various complex tasks in a zero-shot manner, they remain susceptible to jailbreak attacks, particularly the unknown jailbreak attack. To…
How should we evaluate the robustness of language model defenses? Current defenses against jailbreaks and prompt injections (which aim to prevent an attacker from eliciting harmful knowledge or remotely triggering malicious actions,…
Fine-grained Address Space Randomization has been considered as an effective protection against code reuse attacks such as ROP/JOP. However, it only employs a one-time randomization, and such a limitation has been exploited by recent…
Algebraic Manipulation Detection (AMD) Codes detect adversarial noise that is added to a coded message and stored in a storage that is opaque to the adversary. We study AMD codes when the storage can leak up to \rho\log|G| bits of…
Adversarial attack perturbs an image with an imperceptible noise, leading to incorrect model prediction. Recently, a few works showed inherent bias associated with such attack (robustness bias), where certain subgroups in a dataset (e.g.…
Understanding properties of deep neural networks is an important challenge in deep learning. In this paper, we take a step in this direction by proposing a rigorous way of verifying properties of a popular class of neural networks,…
ASIC hash engines are specifically optimized for parallel computations of cryptographic hashes and thus a natural environment for mounting brute-force attacks on hash functions. Two fundamental advantages of ASICs over general purpose…
The search for Boolean functions that can withstand the main crypyographic attacks is essential. In 2008, Carlet and Feng studied a class of functions which have optimal cryptographic properties with the exception of nonlinearity for which…
Presentation Attack Detection (PAD) systems are usually designed independently of the fingerprint verification system. While this can be acceptable for use cases where specific user templates are not predetermined, it represents a missed…
Pre-trained language models of code are now widely used in various software engineering tasks such as code generation, code completion, vulnerability detection, etc. This, in turn, poses security and reliability risks to these models. One…
Facial attributes are emerging soft biometrics that have the potential to reject non-matches, for example, based on mismatching gender. To be usable in stand-alone systems, facial attributes must be extracted from images automatically and…
LoRa provides long-range, energy-efficient communications in Internet of Things (IoT) applications that rely on Low-Power Wide-Area Network (LPWAN) capabilities. Despite these merits, concerns persist regarding the security of LoRa…
We revisit Christol's theorem on algebraic power series in positive characteristic and propose yet another proof for it. This new proof combines several ingredients and advantages of existing proofs, which make it very well-suited for…
Timing-based side and covert channels in processor caches continue to be a threat to modern computers. This work shows for the first time a systematic, large-scale analysis of Arm devices and the detailed results of attacks the processors…
Recent works have shown that the input domain of any machine learning classifier is bound to contain adversarial examples. Thus we can no longer hope to immune classifiers against adversarial examples and instead can only aim to achieve the…
Upcoming certification actions related to the security of machine learning (ML) based systems raise major evaluation challenges that are amplified by the large-scale deployment of models in many hardware platforms. Until recently, most of…
Large language models (LLMs) employ safety mechanisms to prevent harmful outputs, yet these defenses primarily rely on semantic pattern matching. We show that encoding harmful prompts as coherent mathematical problems -- using formalisms…
Adversarial defenses are naturally evaluated on their ability to tolerate adversarial attacks. To test defenses, diverse adversarial attacks are crafted, that are usually described in terms of their evading capability and the L0, L1, L2,…
Due to the adoption of horizontal business models following the globalization of semiconductor manufacturing, the overproduction of integrated circuits (ICs) and the piracy of intellectual properties (IPs) can lead to significant damage to…