Contextual Fine-Tuning of Language Models with Classifier-Driven Content Moderation for Text Generation

In today’s digital age, ensuring the appropriateness of content for children is crucial for their cognitive and emotional development. The rise of automated text generation technologies, such as Large Language Models like LLaMA, Mistral, and Zephyr, has created a pressing need for effective tools to...

Full description

Saved in:
Bibliographic Details
Main Authors: Matan Punnaivanam, Palani Velvizhy
Format: Article
Language:English
Published: MDPI AG 2024-12-01
Series:Entropy
Subjects:
Online Access:https://www.mdpi.com/1099-4300/26/12/1114
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850050896463921152
author Matan Punnaivanam
Palani Velvizhy
author_facet Matan Punnaivanam
Palani Velvizhy
author_sort Matan Punnaivanam
collection DOAJ
description In today’s digital age, ensuring the appropriateness of content for children is crucial for their cognitive and emotional development. The rise of automated text generation technologies, such as Large Language Models like LLaMA, Mistral, and Zephyr, has created a pressing need for effective tools to filter and classify suitable content. However, the existing methods often fail to effectively address the intricate details and unique characteristics of children’s literature. This study aims to bridge this gap by developing a robust framework that utilizes fine-tuned language models, classification techniques, and contextual story generation to generate and classify children’s stories based on their suitability. Employing a combination of fine-tuning techniques on models such as LLaMA, Mistral, and Zephyr, alongside a BERT-based classifier, we evaluated the generated stories against established metrics like ROUGE, METEOR, and BERT Scores. The fine-tuned Mistral-7B model achieved a ROUGE-1 score of 0.4785, significantly higher than the base model’s 0.3185, while Zephyr-7B-Beta achieved a METEOR score of 0.4154 compared to its base counterpart’s score of 0.3602. The results indicated that the fine-tuned models outperformed base models, generating content more aligned with human standards. Moreover, the BERT Classifier exhibited high precision (0.95) and recall (0.97) for identifying unsuitable content, further enhancing the reliability of content classification. These findings highlight the potential of advanced language models in generating age-appropriate stories and enhancing content moderation strategies. This research has broader implications for educational technology, content curation, and parental control systems, offering a scalable approach to ensuring children’s exposure to safe and enriching narratives.
format Article
id doaj-art-4714738971ea48aaa38cc870b2ee7838
institution DOAJ
issn 1099-4300
language English
publishDate 2024-12-01
publisher MDPI AG
record_format Article
series Entropy
spelling doaj-art-4714738971ea48aaa38cc870b2ee78382025-08-20T02:53:19ZengMDPI AGEntropy1099-43002024-12-012612111410.3390/e26121114Contextual Fine-Tuning of Language Models with Classifier-Driven Content Moderation for Text GenerationMatan Punnaivanam0Palani Velvizhy1Department of Computer Science and Engineering, College of Engineering Guindy, Anna University, Chennai 600025, IndiaDepartment of Computer Science and Engineering, College of Engineering Guindy, Anna University, Chennai 600025, IndiaIn today’s digital age, ensuring the appropriateness of content for children is crucial for their cognitive and emotional development. The rise of automated text generation technologies, such as Large Language Models like LLaMA, Mistral, and Zephyr, has created a pressing need for effective tools to filter and classify suitable content. However, the existing methods often fail to effectively address the intricate details and unique characteristics of children’s literature. This study aims to bridge this gap by developing a robust framework that utilizes fine-tuned language models, classification techniques, and contextual story generation to generate and classify children’s stories based on their suitability. Employing a combination of fine-tuning techniques on models such as LLaMA, Mistral, and Zephyr, alongside a BERT-based classifier, we evaluated the generated stories against established metrics like ROUGE, METEOR, and BERT Scores. The fine-tuned Mistral-7B model achieved a ROUGE-1 score of 0.4785, significantly higher than the base model’s 0.3185, while Zephyr-7B-Beta achieved a METEOR score of 0.4154 compared to its base counterpart’s score of 0.3602. The results indicated that the fine-tuned models outperformed base models, generating content more aligned with human standards. Moreover, the BERT Classifier exhibited high precision (0.95) and recall (0.97) for identifying unsuitable content, further enhancing the reliability of content classification. These findings highlight the potential of advanced language models in generating age-appropriate stories and enhancing content moderation strategies. This research has broader implications for educational technology, content curation, and parental control systems, offering a scalable approach to ensuring children’s exposure to safe and enriching narratives.https://www.mdpi.com/1099-4300/26/12/1114large language models (LLMs)supervised fine tuningtext classificationBERTgenerative modelsdeep learning
spellingShingle Matan Punnaivanam
Palani Velvizhy
Contextual Fine-Tuning of Language Models with Classifier-Driven Content Moderation for Text Generation
Entropy
large language models (LLMs)
supervised fine tuning
text classification
BERT
generative models
deep learning
title Contextual Fine-Tuning of Language Models with Classifier-Driven Content Moderation for Text Generation
title_full Contextual Fine-Tuning of Language Models with Classifier-Driven Content Moderation for Text Generation
title_fullStr Contextual Fine-Tuning of Language Models with Classifier-Driven Content Moderation for Text Generation
title_full_unstemmed Contextual Fine-Tuning of Language Models with Classifier-Driven Content Moderation for Text Generation
title_short Contextual Fine-Tuning of Language Models with Classifier-Driven Content Moderation for Text Generation
title_sort contextual fine tuning of language models with classifier driven content moderation for text generation
topic large language models (LLMs)
supervised fine tuning
text classification
BERT
generative models
deep learning
url https://www.mdpi.com/1099-4300/26/12/1114
work_keys_str_mv AT matanpunnaivanam contextualfinetuningoflanguagemodelswithclassifierdrivencontentmoderationfortextgeneration
AT palanivelvizhy contextualfinetuningoflanguagemodelswithclassifierdrivencontentmoderationfortextgeneration