Vall Herard

Title: Vall Herard and Compliance Text Classification

Introduction

Vall Herard is an inventor based in Boston, Massachusetts, and is associated with FMR Corp. His patent work addresses the use of natural language processing to identify potential compliance violations in text.

Latest Patents

Herard is associated with 1 patent: “Classification of Compliance Text Data with Few Shot Natural Language Model.” The patent describes methods and apparatuses for classifying compliance text when little or no training data is available. A server receives structured or unstructured compliance text from a database and applies a trained few-shot natural language inference classification model to sentences in that text. The model identifies whether sentences may contain compliance violations, and the server sends its results to a remote computing device.

Career Highlights

Herard’s listed invention applies few-shot natural language inference to compliance review. This approach is designed to help classify text without relying on extensive task-specific training data.

Collaborations

Herard’s listed coworkers include John McCrae and Jason Matthew Megaro. The available information identifies these professional associations but does not specify their roles in the patent.

Conclusion

Vall Herard’s patent focuses on applying few-shot natural language models to compliance text, offering a method for identifying potential violations with limited training data.