MUMBAI, India, May 29 -- Intellectual Property India has published a patent application (202641053706 A) filed by G V Manoj Kumar; Dr. A. Gautami Latha; Dr. G. Sandhya Devi; Gandalla K V Kankshita; Goljana Kavya; Gandreddy Himasri; and Dasari Melissa Keerthana, Visakhapatnam, Andhra Pradesh, on April 27, for 'ai-powered expense scanner & travel claim analyzer using vision-language models for enterprises.'
Inventor(s) include G V Manoj Kumar; Dr. A. Gautami Latha; Dr. G. Sandhya Devi; Gandalla K V Kankshita; Goljana Kavya; Gandreddy Himasri; and Dasari Melissa Keerthana.
The application for the patent was published on May 29, under issue no. 22/2026.
According to the abstract released by the Intellectual Property India: "AI-powered expense scanner & travel claim analyzer using vision-language models for enterprises is the proposed invention. In an organization auditing plays a vital role, among which calculations of the employees' expenses is the toughest task where the auditors and the employees' ratio are immense which increases the workload on the auditors. This makes the auditing process susceptible to errors. The growing number of expense claims in organizations has created a need for smarter and more efficient verification systems. Manually reviewing receipts and supporting documents takes a lot of time and often lacks consistency, causing delays and operational inefficiencies. Traditional OCR-based solutions can pull text from travel tickets, hotel invoices, and purchase bills, but they do not understand the context or relationships between important details like vendor name, date, category, and total amount. This limitation makes it hard to validate claims accurately or spot issues like duplicate submissions and inflated expenses. AI- Powered Expense Scanner & Travel Claim Analyzer using Vision Language Models (VLM) for Enterprises is used to automate the extraction and validation of expense data from receipts as the reimbursement process. This system reduces manual effort, improves accuracy, and enhances transparency in the auditing process. It demonstrates the usage of Vision-Language Models like PaliGemma and QWEN with model accuracy of 91% in solving real-world challenges and provides a scalable structure for future improvements."
Disclaimer: Curated by HT Syndication.