Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{195039,
        author = {M.A.Reetha Jeyarani and L.Josephine Usha},
        title = {Transformer-Based Deep Learning Framework of the Legal Document Summarization},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {12},
        number = {10},
        pages = {6198-6206},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=195039},
        abstract = {The legal documents are usually written in complicated language, have myriads of references and procedural description which cannot be easily understood by legal professionals and researchers. In this paper, a transformer-based deep learning solution to automated summarizing of legal documents to extract the essential information and to simplify the complex legal document formats is described. It is a structured system that combines a higher level of preprocessing, contextual embedding generation and attention-based transformer architecture to read legal documents and generate concise abstractive summaries. The system is conditioned with legal datasets that are publicly available with court judgments and legislative documents where the model is able to learn contextual relationships in legal discourse. Experimental results indicate that it yields results using ROUGE and semantic similarity testing, and that the summarization accuracy is 96.2, ROUGE-1 evaluation score is 0.92, and ROUGE-2 evaluation score was 0.84. The summaries generated are effective in storing key legal reasoning and saving a lot of space in length of documents. The suggested method will enhance a better reading process and provide a quicker legal document analysis. The findings suggest the usefulness of the transformer-based contextual learning to summarize legal texts. The future evolution can combine domain-adaptive legal knowledge graphs and multi-language summarization functions to improve automated legal information extraction and accessibility even more.},
        keywords = {Summarization In legal documents, transformer model, abstractive summarization, natural language processing, legal text mining, deep learning, contextual embedding.},
        month = {March},
        }