Intellectual Property India Publishes Patent Application for 'Multimodal Fake Content Detection Using Graph Neural Networks And Convolutional Neural Networks' Filed by Vellore Institute Of Technology

MUMBAI, India, Jan. 2 -- Intellectual Property India has published a patent application (202541122238 A) filed by Vellore Institute Of Technology, Vellore, Tamil Nadu, on Dec. 4, 2025, for 'multimodal fake content detection using graph neural networks and convolutional neural networks.'

Inventor(s) include Dr. D. Ajitha; Nishit Dua; and Jeffrey Jeeva Dhas.

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 multimodal fake content detection system comprising a text processing module configured to receive textual content, a Graph Neural Network model configured to analyze the textual content and generate a first authenticity prediction, an audio processing module configured to receive audio content, a Convolutional Neural Network model configured to analyze the audio content and generate a second authenticity prediction, and a fusion module configured to combine the first authenticity prediction and the second authenticity prediction to generate a combined authenticity assessment. The text processing module generates a graph representation from the textual content where nodes represent semantic entities and edges represent contextual relationships between the semantic entities. The audio processing module converts the audio content into spectrogram representations for analysis by the Convolutional Neural Network model."

Disclaimer: Curated by HT Syndication.