We provide baselines and an initial analysis of RepE techniques, showing that they offer simple yet effective solutions for improving our understanding and control of large language models.

Andy Zou. PhD Student, Carnegie Mellon University. Verified email at andrew.cmu.edu - Homepage. ML Safety AI Safety. Articles Cited by Public access. Title. Sort. Sort by citations Sort by year Sort by title. Cited by. Cited by. Year; Measuring …

2023年10月2日 · In this paper, we identify and characterize the emerging area of representation engineering (RepE), an approach to enhancing the transparency of AI systems that draws on insights from cognitive neuroscience.

2023年7月29日 · Andy Zou是CMU计算机科学系的一名一年级博士生，导师是Zico Kolter和Matt Fredrikson。 此前，他在UC伯克利获得了硕士和学士学位，导师是Dawn Song和Jacob Steinhardt。

2023年7月27日 · View a PDF of the paper titled Universal and Transferable Adversarial Attacks on Aligned Language Models, by Andy Zou and 5 other authors

Andy Zou is a multi-disciplinary content creator, director, and tech enthusiast with over a decade of experience in the social-first video content.

2024年8月1日 · This is the official repository for "Universal and Transferable Adversarial Attacks on Aligned Language Models" by Andy Zou, Zifan Wang, Nicholas Carlini, Milad Nasr, J. Zico Kolter, and Matt Fredrikson. Check out our website and demo here.

Andy is a jack-of-all-trades Renaissance guy of creative digital content. A director, cinematographer, editor, writer, and sometimes coder, Andy has worked on hundreds of short and long-form projects over the past decade, establishing a reputation as a go-to guy for turnkey video and content production.

Andy Zou Big Data for Global Employment Dynamics. For this project, I will be developing an open-source machine learning library that makes Natural Language Processing tasks easier.