Towards Consistent Detection of Cognitive Distortions: LLM-Based Annotation and Dataset-Agnostic Evaluation

TL;DR

This paper investigates using Large Language Models, especially GPT-4, as consistent annotators for cognitive distortion detection, proposing a dataset-agnostic evaluation framework to improve reliability and comparability across datasets.

Contribution

It introduces a novel LLM-based annotation approach and a dataset-agnostic evaluation method using Cohen's kappa for subjective NLP tasks.

Findings

GPT-4 achieves Fleiss's Kappa of 0.78 for consistent annotations.

Models trained on LLM-annotated data outperform those trained on human labels.

The proposed framework enables fair cross-dataset comparison of NLP models.

Abstract

Text-based automated Cognitive Distortion detection is a challenging task due to its subjective nature, with low agreement scores observed even among expert human annotators, leading to unreliable annotations. We explore the use of Large Language Models (LLMs) as consistent and reliable annotators, and propose that multiple independent LLM runs can reveal stable labeling patterns despite the inherent subjectivity of the task. Furthermore, to fairly compare models trained on datasets with different characteristics, we introduce a dataset-agnostic evaluation framework using Cohen's kappa as an effect size measure. This methodology allows for fair cross-dataset and cross-study comparisons where traditional metrics like F1 score fall short. Our results show that GPT-4 can produce consistent annotations (Fleiss's Kappa = 0.78), resulting in improved test set performance for models trained on these annotations compared to those trained on human-labeled data. Our findings suggest that LLMs can offer a scalable and internally consistent alternative for generating training data that supports strong downstream performance in subjective NLP tasks.