Style Transfer as Bias Mitigation: Diffusion Models for Synthetic Mental Health Text for Arabic

TL;DR

This paper introduces a diffusion model approach for synthetic Arabic mental health text generation that mitigates gender bias by performing style transfer without pretrained language models, enhancing diversity and fairness.

Contribution

It presents a novel, pretraining-free diffusion-based style transfer method for bias mitigation in low-resource mental health text data, specifically targeting gender imbalance in Arabic.

Findings

High semantic fidelity in generated text

Effective gender style transfer demonstrated

Mitigates bias without pretrained LLMs

Abstract

Synthetic data offers a promising solution for mitigating data scarcity and demographic bias in mental health analysis, yet existing approaches largely rely on pretrained large language models (LLMs), which may suffer from limited output diversity and propagate biases inherited from their training data. In this work, we propose a pretraining-free diffusion-based approach for synthetic text generation that frames bias mitigation as a style transfer problem. Using the CARMA Arabic mental health corpus, which exhibits a substantial gender imbalance, we focus on male-to-female style transfer to augment underrepresented female-authored content. We construct five datasets capturing varying linguistic and semantic aspects of gender expression in Arabic and train separate diffusion models for each setting. Quantitative evaluations demonstrate consistently high semantic fidelity between source and generated text, alongside meaningful surface-level stylistic divergence, while qualitative analysis confirms linguistically plausible gender transformations. Our results show that diffusion-based style transfer can generate high-entropy, semantically faithful synthetic data without reliance on pretrained LLMs, providing an effective and flexible framework for mitigating gender bias in sensitive, low-resource mental health domains.