Skip to yearly menu bar Skip to main content


Poster

Adversarial Representation Engineering: A General Model Editing Framework for Large Language Models

Yihao Zhang · Zeming Wei · Jun Sun · Meng Sun

East Exhibit Hall A-C #4903
[ ]
Wed 11 Dec 11 a.m. PST — 2 p.m. PST

Abstract:

Since the rapid development of Large Language Models (LLMs) has achieved remarkable success, understanding and rectifying their internal complex mechanisms has become an urgent issue. Recent research has attempted to interpret their behaviors through the lens of inner representation. However, developing practical and efficient methods for applying these representations for general and flexible model editing remains challenging. In this work, we explore how to leverage insights from representation engineering to guide the editing of LLMs by deploying a representation sensor as an editing oracle. We first identify the importance of a robust and reliable sensor during editing, then propose an \textbf{A}dversarial \textbf{R}epresentation \textbf{E}ngineering (\textbf{ARE}) framework to provide a unified and interpretable approach for conceptual model editing without compromising baseline performance. Experiments on multiple tasks demonstrate the effectiveness of ARE in various model editing scenarios. Our code and data are available at \url{https://github.com/Zhang-Yihao/Adversarial-Representation-Engineering}.

Live content is unavailable. Log in and register to view live content