Sockpuppetting: Jailbreaking LLMs without Optimization through Output Prefix Injection

As open-weight large language models LLMs increase in capabilities, safeguarding them against malicious prompts and understanding possible attack vectors becomes ever more important. While automated jailbreaking methods like GCG Zou et al., 2023 remain effective, they often require substantial...

DEBIAN-CVE-2018-7648

An issue was discovered in mj2/opjmj2extract.c in OpenJPEG 2.3.0. The output prefix was not checked for length, which could overflow a buffer, when providing a prefix with 50 or more characters on the command line...

CVE-2018-7648

An issue was discovered in mj2/opjmj2extract.c in OpenJPEG 2.3.0. The output prefix was not checked for length, which could overflow a buffer, when providing a prefix with 50 or more characters on the command line...