May 14, 2024, 4:11 a.m. | Bo Hui, Haolin Yuan, Neil Gong, Philippe Burlina, Yinzhi Cao

cs.CR updates on

arXiv:2405.06823v1 Announce Type: new
Abstract: Large Language Models (LLMs) enable a new ecosystem with many downstream applications, called LLM applications, with different natural language processing tasks. The functionality and performance of an LLM application highly depend on its system prompt, which instructs the backend LLM on what task to perform. Therefore, an LLM application developer often keeps a system prompt confidential to protect its intellectual property. As a result, a natural attack, called prompt leaking, is to steal the system …

application applications arxiv attacks backend called cs.lg ecosystem enable language language models large large language model llm llms natural natural language natural language processing performance prompt system task

Sr. Product Manager

@ MixMode | Remote, US

Information Security Engineers

@ D. E. Shaw Research | New York City

Data Privacy Manager

@ EY | Sydney, NSW, AU, 2000

Senior Security Analyst (SOC)

@ Accesa & RaRo | Cluj-Napoca, Romania

Network & Security Engineer

@ Talan | Genève, Switzerland

Manager, Forensics and Litigation Support

@ MNP | Toronto, Ontario, Canada