Enhanced Detection of Conversational Mental Manipulation Through Advanced Prompting Techniques

TL;DR

This paper investigates advanced prompting techniques for detecting conversational mental manipulation, aiming to understand their effectiveness and develop a specialized detection framework.

Contribution

It introduces a novel framework for mental manipulation detection using Chain-of-Thought prompting in zero-shot and few-shot settings, analyzing their performance.

Findings

Advanced prompting techniques may be less effective for complex models without example-based training.

Chain-of-Thought prompting improves detection in simple models.

Performance varies significantly across different prompting strategies.

Abstract

This study presents a comprehensive, long-term project to explore the effectiveness of various prompting techniques in detecting dialogical mental manipulation. We implement Chain-of-Thought prompting with Zero-Shot and Few-Shot settings on a binary mental manipulation detection task, building upon existing work conducted with Zero-Shot and Few- Shot prompting. Our primary objective is to decipher why certain prompting techniques display superior performance, so as to craft a novel framework tailored for detection of mental manipulation. Preliminary findings suggest that advanced prompting techniques may not be suitable for more complex models, if they are not trained through example-based learning.