Pre-Trained Language Models development using Contrastive Framework for Semi-Supervised Fine-Tuning

Author(s)	Mr. Rohit Singh
Country	India
Abstract	The rapid advancements in pre-trained language models (PLMs) have revolutionized natural language processing (NLP), giving performance across diverse tasks. However, their efficacy diminishes in low-resource domains with limited labeled data, where extracting task-specific semantics becomes challenging. This limitation is particularly pronounced in mission-critical applications such as military operations, where the availability of labeled datasets is constrained by security and operational restrictions. To address these challenges, this paper proposes a novel Contrastive Framework for Semi-Supervised Fine-Tuning of PLMs. By integrating contrastive learning with semi-supervised techniques, the framework enables PLMs to effectively leverage both labeled and unlabeled data, enhancing their ability to generalize in low-resource settings. The study focuses on creating a customized, domain-specific language model tailored to the unique linguistic and operational requirements of the Indian Army, addressing critical tasks such as secure communication, multilingual processing, and intelligence analysis.
Keywords	NLP, LLM, machine learning, fine tuning
Field	Engineering
Published In	Volume 16, Issue 3, July-September 2025
Published On	2025-07-15
DOI	https://doi.org/10.71097/IJSAT.v16.i3.6917
Short DOI	https://doi.org/g9s9v6

About IJSAT Fees & Payment Current Issue Publication Archive	Submit Research Paper Track Submission Status Publication Guidelines Publication Ethics Peer Review & Plagiarism	Join as a Reviewer Editors & Reviewers Reviewer Referral Program Get Reviewer Membership Certi.	Website/Journal Policies Usage Policy Content Policies Privacy Policy

Contact Us	Message on WhatsApp	+91-9687-182-185	editor@ijsat.org

International Journal on Science and Technology