May I Have Your Attention? Breaking Fine-Tuning Based Prompt Injection Defenses Using Architecture-Aware Attacks

A popular class of defenses against prompt injection attacks on large language models LLMs relies on fine-tuning the model to separate instructions and data, so that the LLM does not follow instructions that might be present with data. There are several academic systems and production-level...

TuneShield: Mitigating Toxicity in Conversational AI While Fine-Tuning on Untrusted Data

Recent advances in foundation models, such as LLMs, have revolutionized conversational AI. Chatbots are increasingly being developed by customizing LLMs on specific conversational datasets. However, mitigating toxicity during this customization, especially when dealing with untrusted training dat...

SV-LLM: an Agentic Approach for SoC Security Verification Using Large Language Models

Ensuring the security of complex system-on-chips SoCs designs is a critical imperative, yet traditional verification techniques struggle to keep pace due to significant challenges in automation, scalability, comprehensiveness, and adaptability. The advent of large language models LLMs, with their...

Leaner Training, Lower Leakage: Revisiting Memorization in LLM Fine-Tuning with LoRA

Memorization in large language models LLMs makes them vulnerable to data extraction attacks. While pre-training memorization has been extensively studied, fewer works have explored its impact in fine-tuning, particularly for LoRA fine-tuning, a widely adopted parameter-efficient method. In this...

Attack Smarter: Attention-Driven Fine-Grained Webpage Fingerprinting Attacks

Website Fingerprinting WF attacks aim to infer which websites a user is visiting by analyzing traffic patterns, thereby compromising user anonymity. Although this technique has been demonstrated to be effective in controlled experimental environments, it remains largely limited to small-scale...

QGuard:Question-Based Zero-Shot Guard for Multi-Modal LLM Safety

The recent advancements in Large Language ModelsLLMs have had a significant impact on a wide range of fields, from general domains to specialized areas. However, these advancements have also significantly increased the potential for malicious users to exploit harmful and jailbreak prompts for...

MEraser: an Effective Fingerprint Erasure Approach for Large Language Models

Large Language Models LLMs have become increasingly prevalent across various sectors, raising critical concerns about model ownership and intellectual property protection. Although backdoor-based fingerprinting has emerged as a promising solution for model authentication, effective attacks for...

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 ...

Israel Says Iran Is Hacking Security Cameras for Spying

Plus: Ukrainian hackers reportedly knock out a key Russian internet provider, China’s Salt Typhoon hackers claim another victim, and the UK hits 23andMe with a hefty fine over its 2023 data breach...

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...

Differentiation-Based Extraction of Proprietary Data from Fine-Tuned LLMs

The increasing demand for domain-specific and human-aligned Large Language Models LLMs has led to the widespread adoption of Supervised Fine-Tuning SFT techniques. SFT datasets often comprise valuable instruction-response pairs, making them highly valuable targets for potential extraction. This...

Certified Unlearning for Neural Networks

We address the problem of machine unlearning, where the goal is to remove the influence of specific training data from a model upon request, motivated by privacy concerns and regulatory requirements such as the "right to be forgotten." Unfortunately, existing methods rely on restrictive assumptio...

IF-GUIDE: Influence Function-Guided Detoxification of LLMs

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...

CVE-2025-49011 SpiceDB checks involving relations with caveats can result in no permission when permission is expected

SpiceDB is an open source database for storing and querying fine-grained authorization data. Prior to version 1.44.2, on schemas involving arrows with caveats on the arrow’ed relation, when the path to resolve a CheckPermission request involves the evaluation of multiple caveated branches, reques...

Why LLM Safety Guardrails Collapse after Fine-Tuning: a Similarity Analysis between Alignment and Fine-Tuning Datasets

Recent advancements in large language models LLMs have underscored their vulnerability to safety alignment jailbreaks, particularly when subjected to downstream fine-tuning. However, existing mitigation strategies primarily focus on reactively addressing jailbreak incidents after safety guardrail...

Learning to Diagnose Privately: DP-Powered LLMs for Radiology Report Classification

Purpose: This study proposes a framework for fine-tuning large language models LLMs with differential privacy DP to perform multi-abnormality classification on radiology report text. By injecting calibrated noise during fine-tuning, the framework seeks to mitigate the privacy risks associated wit...

Sylva: Tailoring Personalized Adversarial Defense in Pre-Trained Models Via Collaborative Fine-Tuning

Whitepaper called Sylva: Tailoring Personalized Adversarial Defense In Pre-Trained Models Via Collaborative Fine-Tuning...

Video Signature: In-Generation Watermarking for Latent Video Diffusion Models

The rapid development of Artificial Intelligence Generated Content AIGC has led to significant progress in video generation but also raises serious concerns about intellectual property protection and reliable content tracing. Watermarking is a widely adopted solution to this issue, but existing...

Keeping an Eye on LLM Unlearning: the Hidden Risk and Remedy

Although Large Language Models LLMs have demonstrated impressive capabilities across a wide range of tasks, growing concerns have emerged over the misuse of sensitive, copyrighted, or harmful data during training. To address these concerns, unlearning techniques have been developed to remove the...