Intellectual Property India Publishes Patent Application for 'Framework For Zero-Shot Named Entity Recognition And Relation Extraction' Filed by Vellore Institute Of Technology

MUMBAI, India, Jan. 2 -- Intellectual Property India has published a patent application (202541122912 A) filed by Vellore Institute Of Technology, Vellore, Tamil Nadu, on Dec. 5, 2025, for 'framework for zero-shot named entity recognition and relation extraction.'

Inventor(s) include Dr. Bhawana Tyagi; and Shashank Kumar.

The application for the patent was published on Jan. 2, under issue no. 01/2026.

According to the abstract released by the Intellectual Property India: "The present disclosure provides a zero-shot learning framework for named entity recognition and relation extraction. The system includes a data processing module for multi-domain text corpora, a task annotation component generating task-annotated data using zero-shot learning without manually annotated training data, a foundation model training system creating trained models, and an application interface enabling model deployment. The framework incorporates three data sources: GPT-3.5 LLM, 25M Words C4 Subset corpus, and 30B Words Corpus feeding into Task-Annotated Data through zero-shot processing and masked self-training. The component distributes information to parallel foundation model architectures including RoBERTa general foundation model and NuNER task-specific foundation model, generating outputs for medical coding, legal document parsing, and financial news analysis."

Disclaimer: Curated by HT Syndication.