Speaker
Description
The Deep Underground Neutrino Experiment (DUNE) is a long-baseline neutrino experiment aimed at addressing fundamental questions such as the matter-antimatter asymmetry in the universe. Currently, DUNE relies on multiple platforms to store internal documentation, including DocDB, Indico (hosted by Fermilab), and EDMS (hosted by CERN). Retrieving relevant historical information—especially from Indico, which requires navigation through multiple subpages and events—can be particularly challenging. In this talk, I will present efforts to develop a dedicated Large Language Model (LLM) tailored for the DUNE collaboration. This LLM integrates content from various DUNE databases into a unified system. When queried, it not only provides relevant responses but also includes accurate reference links to original sources. The tool is designed to streamline access to DUNE-specific information and has the potential to be extended for efficient data analysis in the future.