MUMBAI, India, June 16 -- Intellectual Property India has published a patent application (202611054401 A) filed by Mr. Ishank; Mr. Kartikey Tyagi; Mr. Krishna Jhunjhunwala; and Mr. Pankaj Pratap Singh, Ghaziabad, Uttar Pradesh, on April 29, for 'sommaire a next.js web application for ai-powered pdf summarisation.'
Inventor(s) include Mr. Ishank; Mr. Kartikey Tyagi; Mr. Krishna Jhunjhunwala; and Mr. Pankaj Pratap Singh.
The application for the patent was published on June 5, under issue no. 23/2026.
According to the abstract released by the Intellectual Property India: "The present invention discloses Sommaire, a web-based AI-powered PDF summarisation platform that addresses the growing challenge of information overload in the digital age. The system integrates pre-trained transformer models, specifically BART (Bidirectional and Auto-Regressive Transformers) and T5 (Text-to-Text Transfer Transformer), with a responsive Next.js frontend and a scalable Vercel-hosted backend, to deliver real-time abstractive summarisation of lengthy PDF documents and web-based articles. Existing text summarisation tools rely heavily on extractive methods that produce semantically disjointed or redundant outputs, and fail to account for the complex, non-linear layouts inherent to PDF documents, including multi-column structures, embedded tables, figures, and footnotes. Sommaire resolves these limitations by deploying abstractive summarisation via the Hugging Face Inference API, enabling the system to paraphrase and reinterpret textual content in a manner consistent with human-level comprehension. The platform achieves factual accuracy and coherence as measured bystandard ROUGE evaluation metrics and demonstrates cross-domain applicabilityspanning academic research, journalism, enterprise reporting, and government documentation. Keywords: Abstractive Summarisation, Transformer Models, BART, T5, Hugging Face, Next.js, PDF Processing, Natural Language Processing, Web Application, Vercel Deployment."
Disclaimer: Curated by HT Syndication.