InverTune: Removing Backdoors from Multimodal Contrastive Learning Models Via Trigger Inversion and Activation Tuning

Multimodal contrastive learning models like CLIP have demonstrated remarkable vision-language alignment capabilities, yet their vulnerability to backdoor attacks poses critical security risks. Attackers can implant latent triggers that persist through downstream tasks, enabling malicious control ...

Smart Buildings Energy Consumption Forecasting Using Adaptive Evolutionary Ensemble Learning Models

Smart buildings are gaining popularity because they can enhance energy efficiency, lower costs, improve security, and provide a more comfortable and convenient environment for building occupants. A considerable portion of the global energy supply is consumed in the building sector and plays a...

Shrinking the Generation-Verification Gap with Weak Verifiers

Verifiers can improve language model capabilities by scoring and ranking responses from generated candidates. Currently, high-quality verifiers are either unscalable e.g., humans or limited in utility e.g., tools like Lean. While LM judges and reward models have become broadly useful as...

Differential Privacy in Machine Learning: from Symbolic AI to LLMs

Machine learning models should not reveal particular information that is not otherwise accessible. Differential privacy provides a formal framework to mitigate privacy risks by ensuring that the inclusion or exclusion of any single data point does not significantly alter the output of an algorith...

VulStamp: Vulnerability Assessment Using Large Language Model

Although modern vulnerability detection tools enable developers to efficiently identify numerous security flaws, indiscriminate remediation efforts often lead to superfluous development expenses. This is particularly true given that a substantial portion of detected vulnerabilities either possess...

Universal Jailbreak Suffixes Are Strong Attention Hijackers

We study suffix-based jailbreaks$\unicodex2013$a powerful family of attacks against large language models LLMs that optimize adversarial suffixes to circumvent safety alignment. Focusing on the widely used foundational GCG attack Zou et al., 2023, we observe that suffixes vary in efficacy: some...

NAP-Tuning: Neural Augmented Prompt Tuning for Adversarially Robust Vision-Language Models

Vision-Language Models VLMs such as CLIP have demonstrated remarkable capabilities in understanding relationships between visual and textual data through joint embedding spaces. Despite their effectiveness, these models remain vulnerable to adversarial attacks, particularly in the image modality,...

SecFwT: Efficient Privacy-Preserving Fine-Tuning of Large Language Models Using Forward-Only Passes

Large language models LLMs have transformed numerous fields, yet their adaptation to specialized tasks in privacy-sensitive domains, such as healthcare and finance, is constrained by the scarcity of accessible training data due to stringent privacy requirements. Secure multi-party computation...

CipherMind: the Longest Codebook in the World

In recent years, the widespread application of large language models has inspired us to consider using inference for communication encryption. We therefore propose CipherMind, which utilizes intermediate results from deterministic fine-tuning of large model inferences as transmission content. The...

LLM-Powered Intent-Based Categorization of Phishing Emails

Phishing attacks remain a significant threat to modern cybersecurity, as they successfully deceive both humans and the defense mechanisms intended to protect them. Traditional detection systems primarily focus on email metadata that users cannot see in their inboxes. Additionally, these systems...

RAS-Eval: a Comprehensive Benchmark for Security Evaluation of LLM Agents in Real-World Environments

The rapid deployment of Large language model LLM agents in critical domains like healthcare and finance necessitates robust security frameworks. To address the absence of standardized evaluation benchmarks for these agents in dynamic environments, we introduce RAS-Eval, a comprehensive security...

Don't Throw the Baby out with the Bathwater: How and Why Deep Learning for ARC

The Abstraction and Reasoning Corpus ARC-AGI presents a formidable challenge for AI systems. Despite the typically low performance on ARC, the deep learning paradigm remains the most effective known strategy for generating skillful state-of-the-art neural networks NN across varied modalities and...

LASA: Enhancing SoC Security Verification with LLM-Aided Property Generation

Ensuring the security of modern System-on-Chip SoC designs poses significant challenges due to increasing complexity and distributed assets across the intellectual property IP blocks. Formal property verification FPV provides the capability to model and validate design behaviors through security...

Detecting Hard-Coded Credentials in Software Repositories Via LLMs

Software developers frequently hard-code credentials such as passwords, generic secrets, private keys, and generic tokens in software repositories, even though it is strictly advised against due to the severe threat to the security of the software. These credentials create attack surfaces...

Weakest Link in the Chain: Security Vulnerabilities in Advanced Reasoning Models

The introduction of advanced reasoning capabilities have improved the problem-solving performance of large language models, particularly on math and coding benchmarks. However, it remains unclear whether these reasoning models are more or less vulnerable to adversarial prompt attacks than their...

AIRTBench: Measuring Autonomous AI Red Teaming Capabilities in Language Models

We introduce AIRTBench, an AI red teaming benchmark for evaluating language models' ability to autonomously discover and exploit Artificial Intelligence and Machine Learning AI/ML security vulnerabilities. The benchmark consists of 70 realistic black-box capture-the-flag CTF challenges from the...

From LLMs to MLLMs to Agents: a Survey of Emerging Paradigms in Jailbreak Attacks and Defenses within LLM Ecosystem

Large language models LLMs are rapidly evolving from single-modal systems to multimodal LLMs and intelligent agents, significantly expanding their capabilities while introducing increasingly severe security risks. This paper presents a systematic survey of the growing complexity of jailbreak...

International Security Applications of Flexible Hardware-Enabled Guarantees

As AI capabilities advance rapidly, flexible hardware-enabled guarantees flexHEGs offer opportunities to address international security challenges through comprehensive governance frameworks. This report examines how flexHEGs could enable internationally trustworthy AI governance by establishing...

Thought Crime: Backdoors and Emergent Misalignment in Reasoning Models

Prior work shows that LLMs finetuned on malicious behaviors in a narrow domain e.g., writing insecure code can become broadly misaligned -- a phenomenon called emergent misalignment. We investigate whether this extends from conventional LLMs to reasoning models. We finetune reasoning models on...

A Comprehensive Survey on Underwater Acoustic Target Positioning and Tracking: Progress, Challenges, and Perspectives

Underwater target tracking technology plays a pivotal role in marine resource exploration, environmental monitoring, and national defense security. Given that acoustic waves represent an effective medium for long-distance transmission in aquatic environments, underwater acoustic target tracking h...