all InfoSec News
CleanGen: Mitigating Backdoor Attacks for Generation Tasks in Large Language Models
June 19, 2024, 4:19 a.m. | Yuetai Li, Zhangchen Xu, Fengqing Jiang, Luyao Niu, Dinuka Sahabandu, Bhaskar Ramasubramanian, Radha Poovendran
cs.CR updates on arXiv.org arxiv.org
Abstract: The remarkable performance of large language models (LLMs) in generation tasks has enabled practitioners to leverage publicly available models to power custom applications, such as chatbots and virtual assistants. However, the data used to train or fine-tune these LLMs is often undisclosed, allowing an attacker to compromise the data and inject backdoors into the models. In this paper, we develop a novel inference time defense, named CleanGen, to mitigate backdoor attacks for generation tasks in …
applications arxiv attacks backdoor backdoor attacks chatbots cs.ai cs.cr custom custom applications data language language models large llms performance power remarkable train virtual
More from arxiv.org / cs.CR updates on arXiv.org
Jobs in InfoSec / Cybersecurity
Ground Systems Engineer - Evolved Strategic SATCOM (ESS)
@ The Aerospace Corporation | Los Angeles AFB
Policy and Program Analyst
@ Obsidian Solutions Group | Rosslyn, VA, US
Principal Network Engineering
@ CVS Health | Work At Home-California
Lead Software Engineer
@ Rapid7 | NIS Belfast
Software Engineer II - Java
@ Rapid7 | NIS Belfast
Senior Software Engineer
@ Rapid7 | NIS Belfast