Working on machine translation for e-commerce to make global shopping easier and more accessible.
Research and development in information extraction from business documents. Developed models to extract data into lists and tables from long documents, and built a dataset management system to streamline data preparation and organization.
Research and development in information extraction from business documents. Developed models for extracting data from structured (e.g., lists, tables), unstructured (raw text), and graphical elements (e.g., signatures, stamps, checkboxes), prepared training, testing, and validation datasets, and conducted experiments to evaluate and optimize model performance.