Permissioned LLMs: Enforcing Access Control in Large Language Models

In enterprise settings, organizational data is segregated, siloed and carefully protected by elaborate access control frameworks. These access control structures can completely break down if an LLM fine-tuned on the siloed data serves requests, for downstream tasks, from individuals with disparat...

Cryptography from Lossy Reductions: Towards OWFs from ETH, and Beyond

One-way functions OWFs form the foundation of modern cryptography, yet their unconditional existence remains a major open question. In this work, we study this question by exploring its relation to lossy reductions, i.e., reductions$R$ for which it holds that $IX;RX \ll n$ for all distributions$X...

The Feasibility of Topic-Based Watermarking on Academic Peer Reviews

Large language models LLMs are increasingly integrated into academic workflows, with many conferences and journals permitting their use for tasks such as language refinement and literature summarization. However, their use in peer review remains prohibited due to concerns around confidentiality...

TrojanStego: Your Language Model Can Secretly Be a Steganographic Privacy Leaking Agent

As large language models LLMs become integrated into sensitive workflows, concerns grow over their potential to leak confidential information. We propose TrojanStego, a novel threat model in which an adversary fine-tunes an LLM to embed sensitive context information into natural-looking outputs v...

CVE-2024-38361

Spicedb is an Open Source, Google Zanzibar-inspired permissions database to enable fine-grained authorization for customer applications. Use of an exclusion under an arrow that has multiple resources may resolve to NOPERMISSION when permission is expected. If the resource exists under multiple...

CVE-2024-46989

spicedb is an Open Source, Google Zanzibar-inspired permissions database to enable fine-grained authorization for customer applications. Multiple caveats over the same indirect subject type on the same relation can result in no permission being returned when permission is expected. If the resourc...

CVE-2022-30757

Improper authorization in isemtelephony prior to SMR Jul-2022 Release 1 allows attacker to obtain CID without ACCESSFINELOCATION permission...

CVE-2021-1887

An assertion can be reached in the WLAN subsystem while using the Wi-Fi Fine Timing Measurement protocol in Snapdragon Wired Infrastructure and Networking...

Mitigating Fine-Tuning Risks in LLMs Via Safety-Aware Probing Optimization

The significant progress of large language models LLMs has led to remarkable achievements across numerous applications. However, their ability to generate harmful content has sparked substantial safety concerns. Despite the implementation of safety alignment techniques during the pre-training...

Backdoor Cleaning without External Guidance in MLLM Fine-Tuning

Multimodal Large Language Models MLLMs are increasingly deployed in fine-tuning-as-a-service FTaaS settings, where user-submitted datasets adapt general-purpose models to downstream tasks. This flexibility, however, introduces serious security risks, as malicious fine-tuning can implant backdoors...

CTRAP: Embedding Collapse Trap to Safeguard Large Language Models from Harmful Fine-Tuning

Fine-tuning-as-a-service, while commercially successful for Large Language Model LLM providers, exposes models to harmful fine-tuning attacks. As a widely explored defense paradigm against such attacks, unlearning attempts to remove malicious knowledge from LLMs, thereby essentially preventing th...

Does Low Rank Adaptation Lead to Lower Robustness against Training-Time Attacks?

Low rank adaptation LoRA has emerged as a prominent technique for fine-tuning large language models LLMs thanks to its superb efficiency gains over previous methods. While extensive studies have examined the performance and structural properties of LoRA, its behavior upon training-time attacks...

Safe Delta: Consistently Preserving Safety When Fine-Tuning LLMs on Diverse Datasets

Large language models LLMs have shown great potential as general-purpose AI assistants across various domains. To fully leverage this potential in specific applications, many companies provide fine-tuning API services, enabling users to upload their own data for LLM customization. However,...

DataSentinel: a Game-Theoretic Detection of Prompt Injection Attacks

LLM-integrated applications and agents are vulnerable to prompt injection attacks, where an attacker injects prompts into their inputs to induce attacker-desired outputs. A detection method aims to determine whether a given input is contaminated by an injected prompt. However, existing detection...

Analysing Safety Risks in LLMs Fine-Tuned with Pseudo-Malicious Cyber Security Data

The integration of large language models LLMs into cyber security applications presents significant opportunities, such as enhancing threat analysis and malware detection, but can also introduce critical risks and safety concerns, including personal data leakage and automated generation of new...

Private LoRA Fine-Tuning of Open-Source LLMs with Homomorphic Encryption

Preserving data confidentiality during the fine-tuning of open-source Large Language Models LLMs is crucial for sensitive applications. This work introduces an interactive protocol adapting the Low-Rank Adaptation LoRA technique for private fine-tuning. Homomorphic Encryption HE protects the...

Israeli NSO Group Fined $168M for Pegasus Spyware Attack on WhatsApp

US jury orders NSO Group to pay $168M to WhatsApp and Meta over Pegasus spyware use in 2019…...

The Steganographic Potentials of Language Models

The potential for large language models LLMs to hide messages within plain text steganography poses a challenge to detection and thwarting of unaligned AI agents, and undermines faithfulness of LLMs reasoning. We explore the steganographic capabilities of LLMs fine-tuned via reinforcement learnin...

LLMs' Suitability for Network Security: a Case Study of STRIDE Threat Modeling

Artificial Intelligence AI is expected to be an integral part of next-generation AI-native 6G networks. With the prevalence of AI, researchers have identified numerous use cases of AI in network security. However, there are almost nonexistent studies that analyze the suitability of Large Language...

Backdoor Attacks against Patch-Based Mixture of Experts

As Deep Neural Networks DNNs continue to require larger amounts of data and computational power, Mixture of Experts MoE models have become a popular choice to reduce computational complexity. This popularity increases the importance of considering the security of MoE architectures. Unfortunately,...