Patently Limited

Benjamin James Hoyle

Jerome Spaargaren

User-input unstructured text data describing one or more entities is received, at least one of which is associated with the figure. The unstructured text data is processed to generate corresponding structured text data, and the structured text data is processed to generate a set of parts. A subset of the set of parts identifying parts that are present in the figure is determined and prompt data is formulated for a neural network large language model to generate description text corresponding to the figure. The formulating of the prompt data includes deriving description data associated with the figure from the user-input unstructured text data, deriving part data corresponding to the subset of the set of parts for the figure. The prompt data is sent to the large language model, and in return description text data is received for the figure from the large language model.

Growing community of inventors

Bath, United Kingdom

Benjamin James Hoyle