Model Editing as a Double-Edged Sword: Steering Agent Ethical Behavior Toward Beneficence or Harm

TL;DR

This paper introduces Behavior Editing, a method for precisely steering large language model-based agents' ethical behaviors, supported by a new benchmark, BehaviorBench, to evaluate both beneficial and harmful behavior modifications.

Contribution

The paper presents Behavior Editing as a novel approach for controlling agent ethics and introduces BehaviorBench, a comprehensive benchmark for evaluating behavior editing in complex scenarios.

Findings

Behavior Editing can effectively steer agents toward desired moral behaviors.

It enables both local scenario-specific and global moral alignment shifts.

BehaviorBench validates the versatility and risks of behavior editing across models.

Abstract

Agents based on Large Language Models (LLMs) have demonstrated strong capabilities across a wide range of tasks. However, deploying LLM-based agents in high-stakes domains comes with significant safety and ethical risks. Unethical behavior by these agents can directly result in serious real-world consequences, including physical harm and financial loss. To efficiently steer the ethical behavior of agents, we frame agent behavior steering as a model editing task, which we term Behavior Editing. Model editing is an emerging area of research that enables precise and efficient modifications to LLMs while preserving their overall capabilities. To systematically study and evaluate this approach, we introduce BehaviorBench, a multi-tier benchmark grounded in psychological moral theories. This benchmark supports both the evaluation and editing of agent behaviors across a variety of scenarios, with each tier introducing more complex and ambiguous scenarios. We first demonstrate that Behavior Editing can dynamically steer agents toward the target behavior within specific scenarios. Moreover, Behavior Editing enables not only scenario-specific local adjustments but also more extensive shifts in an agent's global moral alignment. We demonstrate that Behavior Editing can be used to promote ethical and benevolent behavior or, conversely, to induce harmful or malicious behavior. Through extensive evaluations of agents built on frontier LLMs, BehaviorBench validates the effectiveness of behavior editing across a wide range of models and scenarios. Our findings offer key insights into a new paradigm for steering agent behavior, highlighting both the promise and perils of Behavior Editing.