logo

For the Inventor, By the Inventor

Tel Aviv, Israel

Ofir Ben Shoham

This inventor holds 16 USPTO granted patents. Top assignee: Intuit, Inc.. Active years: 2026.

USPTO Granted Patents = 16

% Patents Active = 100.0

Average Co-Inventor Count = 4.0

ph-index = 1


Company Filing History:

goldMedal16 out of 2,528 
 
Intuit, Inc.
 patents

Years Active: 2026

Loading Chart...
16 patents (USPTO):Explore Patents
1. 12737551 - Computer-aware caching for low latency LLM serving (2026-09-15),
2. 12725089 - Speculative retrieval proxy for retrieval augmented generation (2026-09-01),
3. 12712733 - PARITY TOKEN FOR ERROR-LOCALIZED BLOCK DECODING (2026-08-18),
4. 12705462 - Decoder architecture for early-stop automated retrieval augmented generation (2026-08-11),
5. 12699841 - Importance aware selective correction for speculative decoding in large models (2026-08-04),
6. 12694301 - Language model training for direct preference optimization and improved language model (2026-07-28),
7. 12694213 - Entropy-based set block decoding utilizing an adaptive entropy threshold (2026-07-28),
8. 12688185 - Dynamic graph memory for agents (2026-07-21),
9. 12682248 - Direct preference optimization data filtering via answer similarity using language model embeddings (2026-07-14),
10. 12670406 - Confidence-based reward for group relative policy optimization in language models (2026-06-30),
11. 12670326 - Pivotal token search by perplexity (2026-06-30),
12. 12645768 - Speculative textual watermarking in machine learning (2026-06-02),
13. 12639352 - Systems and methods for load-aware context gating for retrieval-augmented generation (2026-05-26),
14. 12614031 - Incremental context composition for latency reduction in language processing machine learning models (2026-04-28),
15. 12608622 - Replay buffer integration for group-based relative policy optimization in machine learning (2026-04-21),

Title: Ofir Ben Shoham: Innovator in Language Model Efficiency

Introduction

Ofir Ben Shoham is a notable inventor based in Petah Tikva, Israel. He has made significant contributions to the field of computational linguistics, particularly in enhancing the efficiency of language models. His innovative approach has led to the development of a unique patent that addresses computational challenges in language processing.

Latest Patents

Ofir Ben Shoham holds a patent for a "Capability aware language model with improved computational efficiency." This method involves receiving input to a language model that initially has a certain computational efficiency. During processing, the model outputs a first boundary token in a selected execution phase. Before this phase, the model is modified using a designated pruning weight mask. It also outputs a second boundary token in a subsequent execution phase, modified by a successive pruning weight mask. This innovative approach ensures that the final computational efficiency of the language model exceeds the initial efficiency, allowing for the generation of multiple tokens during execution phases.

Career Highlights

Ofir is currently employed at Intuit, Inc., where he applies his expertise in language models to develop advanced computational solutions. His work at Intuit reflects his commitment to innovation and efficiency in technology.

Collaborations

Ofir collaborates with Matan Vetzler, contributing to the advancement of their projects and enhancing the overall impact of their work in the field.

Conclusion

Ofir Ben Shoham's contributions to language model efficiency exemplify the importance of innovation in technology. His patent and work at Intuit, Inc. highlight his role as a leading inventor in this domain.

Profile summary based on public USPTO records.
Please report any incorrect information to [email protected]
Loading…