Jonathan Herzig

Title: Innovations by Jonathan Herzig

Introduction

Jonathan Herzig is an accomplished inventor based in Tel Aviv, Israel. He has made significant contributions to the field of language models, particularly in the area of reasoning over tables. With a total of 2 patents, his work is paving the way for advancements in artificial intelligence and natural language processing.

Latest Patents

Herzig's latest patents focus on systems and methods for training language models to reason over tables. These innovations involve pre-training and fine-tuning neural-network-based language models to reason directly over tables without generating logical forms. In some examples, a language model can be pre-trained using masked-language modeling tasks that are synthetically generated from tables pulled from a knowledge corpus. Furthermore, the language model may be further pre-trained using pairs of counterfactual statements generated from those tables, and/or one or more statements that compare selected data from those tables. The fine-tuning process allows for examples that include only a question, an answer, and a table, enabling fine-tuning examples to be harvested directly from existing benchmark datasets or synthetically generated.

Career Highlights

Jonathan Herzig is currently employed at Google Inc., where he continues to innovate and develop cutting-edge technologies. His work is instrumental in enhancing the capabilities of language models, making them more efficient and effective in reasoning tasks.

Collaborations

Some of his notable coworkers include Thomas Müller and Francesco Piccinno, who contribute to the collaborative environment that fosters innovation at Google Inc.

Conclusion

Jonathan Herzig's contributions to the field of language models are significant and impactful. His patents reflect a deep understanding of artificial intelligence and its applications in reasoning over complex data structures.