9 papers
Faithful Dual-constrained Erasure for Robust LLM Safety Alignment
Jiaqing Li, Shide Zhou, Zhibo Zhang +3
Machine unlearning has emerged as a crucial mechanism for removing hazardous knowledge and enforcing safety alignment in Large Language Models (LLMs). However, recent studies revea…
QMFOL: Benchmarking Large Language Model Reasoning via Quantifiable Monadic First-Order Logic Test Case Generation
Xinyi Zheng, Ling Shi, Tianlong Yu +3
Large Language Models (LLMs) have made significant progress in reasoning, particularly in deductive reasoning, which is crucial for high-stakes decision-making. As models improve,…
SensingAgents: A Multi-Agent Collaborative Framework for Robust IMU Activity Recognition
Naiyu Zheng, Tianlong Yu, Haochen Yin +3
Human Activity Recognition (HAR) using Inertial Measurement Unit (IMU) sensors is a cornerstone of mobile health, smart environments, and human-computer interaction. However, curre…
PhySE: A Psychological Framework for Real-Time AR-LLM Social Engineering Attacks
Tianlong Yu, Yang Yang, Ziyi Zhou +5
The emerging threat of AR-LLM-based Social Engineering (AR-LLM-SE) attacks (e.g. SEAR) poses a significant risk to real-world social interactions. In such an attack, a malicious ac…
UNSEEN: A Cross-Stack LLM Unlearning Defense against AR-LLM Social Engineering Attacks
Tianlong Yu, Yang Yang, Xiao Luo +6
Emerging AR-LLM-based Social Engineering attack (e.g., SEAR) is at the edge of posing great threats to real-world social life. In such AR-LLM-SE attack, the attacker can leverage A…
When Safe Models Merge into Danger: Exploiting Latent Vulnerabilities in LLM Fusion
Jiaqing Li, Zhibo Zhang, Shide Zhou +3
Model merging has emerged as a powerful technique for combining specialized capabilities from multiple fine-tuned LLMs without additional training costs. However, the security impl…
REFN: A Reinforcement-Learning-From-Network Framework against 1-day/n-day Exploitations
Tianlong Yu, Lihong Liu, Ziyi Zhou +3
The exploitation of 1 day or n day vulnerabilities poses severe threats to networked devices due to massive deployment scales and delayed patching (average Mean Time To Patch excee…
SEAR: A Multimodal Dataset for Analyzing AR-LLM-Driven Social Engineering Behaviors
Tianlong Yu, Chenghang Ye, Zheyu Yang +8
The SEAR Dataset is a novel multimodal resource designed to study the emerging threat of social engineering (SE) attacks orchestrated through augmented reality (AR) and multimodal…
On the Feasibility of Using MultiModal LLMs to Execute AR Social Engineering Attacks
Ting Bi, Chenghang Ye, Zheyu Yang +8
Augmented Reality (AR) and Multimodal Large Language Models (LLMs) are rapidly evolving, providing unprecedented capabilities for human-computer interaction. However, their integra…