Jonathan Larson

소개

Jonathan Larson is a Senior Principal Data Architect at Microsoft Research working in Special Projects.  He currently leads a research team focused on the intersection of graph machine learning, LLM memory representations, and LLM orchestration.  Some of his group’s recent research include:

• GraphRAG – A method for creating LLM generated knowledge graphs and then subsequently using them for information retrieval.
  • July 2024 Blog: GraphRAG: Unlocking LLM discovery on narrative private data
  • February 2024 Blog: GraphRAG: New tool for complex data discovery now on GitHub
  • Arxiv: From Local to Global: A Graph RAG Approach to Query-Focused Summarization
  • Github: GraphRAG GitHub (opens in new tab)
• Organizational Science – Using graph machine learning to better understand human behaviors in large organizational settings.
  • WorkLab: Patterns hidden inside the org chart
  • Management Science (Forthcoming): Dynamic Silos: Modularity in intra-organizational communication networks before and during the Covid-19 pandemic
  • Video: Developer Tech Minutes: Organizational Analytics
  • Harvard Business Review: What a Year of WFH Has Done to Our Relationships at Work (opens in new tab)
  • WorkLab: The Next Great Disruption is Hybrid Work – Are we Ready?
  • Blog: Advancing organizational science using network machine learning to measure innovation in the workplace
  • Viva Research: Uncovering resilience: Measuring organizational networks during crisis (opens in new tab)
  • Viva Research: Toward resilience: Adapting to crisis through the lens of organizational networks (opens in new tab)
  • Arxiv: Workgroup Mapping: Visual Analysis of Collaboration Culture
•  Cybersecurity – Using graph machine learning and LLMs for cyber-security applications
  • A graph embedding approach to user behavior anomaly detection
  • Binary code summarization: Benchmarking ChatGPT/GPT-4 and other large language models 
  • Using web scale graph analytics to counter technical support scams 
  • A graph based framework for reducing false positives in authentication alerts in security systems

 His research has led to shipping new features in Bing, Viva, PowerBI.  He also shipped new tools to combat tech fraud.  Many of the supporting libraries have been open sourced in collaboration on GitHub (E.g. graphrag (opens in new tab), graspologic, (opens in new tab)showwhy (opens in new tab)).  Prior to joining Microsoft, Jonathan was Chief Scientist and Technical Fellow at Sotera Defense Solutions on assignment to DARPA, and led a variety of research across several programs.  Jonathan has also led large-scale data science efforts at Google, Zillow, and the US Army.  Early in his career, he also worked several startups and incubators.