Harnessing open language models: a systematic literature review unleashing AI's potential for a smarter future

Abstract

This study provides a systematic literature review (SLR) on Open Large Language Models (OLLM), which are large-scale natural language processing (NLP) models with accessible source code, configuration, and training data for the community. Recent advances in supervised and unsupervised learning techniques have improved the accuracy and contextual capabilities of OLLMs, enabling advanced applications in conversational interaction and long-text analysis. This research explored the applications and socioeconomic impacts of OLLMs in various industries, such as healthcare, education, and business management, demonstrating how these models optimize the efficiency and personalization of different processes. The study also addresses the ethical and operational challenges associated with OLLMs, such as bias management, data privacy and security, decision-making transparency, and technological dependency. Strategies are proposed to mitigate these issues, including regular ethics audits and the adoption of explainable AI frameworks. Finally, the study emphasizes the importance of maintaining a balance between OLLMs and human skills, the need for robust governance frameworks to ensure the ethical and legal operation of these models, and the promotion of continuous innovation to expand their capabilities for a positive and lasting impact on society.