2 papers
cs.CR2026
ICON: Intent-Context Coupling for Efficient Multi-Turn Jailbreak Attack
Xingwei Lin, Wenhao Lin, Sicong Cao +4
Multi-turn jailbreak attacks have emerged as a critical threat to Large Language Models (LLMs), bypassing safety mechanisms by progressively constructing adversarial contexts from…
cs.SE2024
Hyperion: Unveiling DApp Inconsistencies using LLM and Dataflow-Guided Symbolic Execution
Shuo Yang, Xingwei Lin, Jiachi Chen +5
The rapid advancement of blockchain platforms has significantly accelerated the growth of decentralized applications (DApps). Similar to traditional applications, DApps integrate f…