PPL
Arthur Gervais
Papers (6)
SCDBench: A Benchmark for LLM-Based Smart Contract DecompilersMeasuring Safety Alignment Effects in Autonomous Security AgentsBenchmarking Mythos-Linked Bug RediscoveryAblating Safety: Mechanisms for Removing Alignment in Language Models for Security ApplicationsCrackMeBench: Binary Reverse Engineering for AgentsPatch2Vuln: Agentic Reconstruction of Vulnerabilities from Linux Distribution Binary Patches