all AI news
Attacking LLM Watermarks by Exploiting Their Strengths
Feb. 27, 2024, 5:43 a.m. | Qi Pang, Shengyuan Hu, Wenting Zheng, Virginia Smith
cs.LG updates on arXiv.org arxiv.org
Abstract: Advances in generative models have made it possible for AI-generated text, code, and images to mirror human-generated content in many applications. Watermarking, a technique that aims to embed information in the output of a model to verify its source, is useful for mitigating misuse of such AI-generated content. However, existing watermarking schemes remain surprisingly susceptible to attack. In particular, we show that desirable properties shared by existing LLM watermarking systems such as quality preservation, robustness, …
abstract advances ai-generated content ai-generated text applications arxiv code cs.cl cs.cr cs.lg embed generated generative generative models human images information llm misuse text type verify watermarking watermarks
More from arxiv.org / cs.LG updates on arXiv.org
Jobs in AI, ML, Big Data
Founding AI Engineer, Agents
@ Occam AI | New York
AI Engineer Intern, Agents
@ Occam AI | US
AI Research Scientist
@ Vara | Berlin, Germany and Remote
Data Architect
@ University of Texas at Austin | Austin, TX
Data ETL Engineer
@ University of Texas at Austin | Austin, TX
Sr. Software Development Manager, AWS Neuron Machine Learning Distributed Training
@ Amazon.com | Cupertino, California, USA