This inventor holds 1 USPTO granted patent and 1 published patent application. Top assignee: Sap Se. Active years: 2026.
Data source: USPTO (United States Patent and Trademark Office) public patent records. Weekly synchronization. How IDiyas builds this profile
Company Filing History:
Years Active: 2026
Title: Abhinandan Padhi and His Patent on Entity Matching
Introduction
Abhinandan Padhi is an inventor associated with Singapore. His patent work focuses on making explanations from entity-matching models more informative through dynamic tokenization.
Latest Patents
Padhi is associated with 1 patent: “Enhanced Model Explanations Using Dynamic Tokenization for Entity Matching Models.” The invention describes methods and systems that process query and target entity data using character-level embeddings and sub-word-level tokens. An attention machine-learning model generates embeddings and attention scores that indicate the relative importance of tokens in a predicted match. The system then uses the attention matrix to produce an explanation for the match.
Career Highlights
Padhi is associated with SAP SE. His patent reflects work at the intersection of machine learning, entity matching, and model explainability.
Collaborations
Sundeep Gullapalli and Rajesh Vellore Arumugam are identified as Padhi’s coworkers. Their association provides context for the professional environment connected with his work.
Conclusion
Abhinandan Padhi’s patent addresses how dynamic tokenization and attention-based analysis can help explain entity-matching predictions.
Data source: USPTO (United States Patent and Trademark Office) public patent records. Weekly synchronization. How IDiyas builds this profile