FERRET: Framework for Expansion Reliant Red Teaming

TL;DR

FERRET is a comprehensive automated red teaming framework that enhances adversarial multi-modal conversations through horizontal, vertical, and meta expansions, outperforming existing methods.

Contribution

The paper introduces FERRET, a novel multi-faceted framework that systematically improves adversarial conversation generation via multiple expansion strategies.

Findings

FERRET generates more effective adversarial conversations.

FERRET outperforms existing red teaming approaches.

Multi-modal attack strategies are effectively discovered during conversations.

Abstract

We introduce a multi-faceted automated red teaming framework in which the goal is to generate multi-modal adversarial conversations that would break a target model and introduce various expansions that would result in more effective and efficient adversarial conversations. The introduced expansions include: 1. Horizontal expansion in which the goal is for the red team model to self-improve and generate more effective conversation starters that would shape a conversation. 2. Vertical expansion in which the goal is to take these conversation starters that are discovered in the horizontal expansion phase and expand them into effective multi-modal conversations and 3. Meta expansion in which the goal is for the red team model to discover more effective multi-modal attack strategies during the course of a conversation. We call our framework FERRET (Framework for Expansion Reliant Red Teaming) and compare it with various existing automated red teaming approaches. In our experiments, we demonstrate the effectiveness of FERRET in generating effective multi-modal adversarial conversations and its superior performance against existing state of the art approaches.