XMM-GPT
Description
XMM-GPT is a domain-specialized AI assistant built by fine-tuning Google's Open-Source FLAN family of LLMs through transfer learning on up-to-date XMM-Newton Documentation. The model is then integrated into a retrieval-augmented generation (RAG) pipeline to ensure that responses are grounded in mission documentation. This reduces model hallucinations and introduces transparency in its decision making process by relaying where in the documentation it got its answer. XMM-GPT acts as an interactive research aide, supporting a wide-range of users (students, researchers, teachers, PIs) in learning about XMM-Newton and its software the SAS.
Detailed example
Instructional Responses (step-by-step guides for SAS tasks), Code Assistance, Explanations/Summarizations of XMM relevant information, Grounded Answers (links to documentation that holds the answer)
AI / analytics pattern
Classical/Predictive Machine Learning: Models trained on data to make predictions or classifications based on identified patterns or relationships.
Automation level / stage
a) Pre-deployment – The use case is in a development or acquisition status.
Expected benefit
Lower entry barriers for new users learning about XMM-Newton and its workflows. Coding assistant that helps to create scripts, explain SAS tasks, and point users to what is causing their coding bugs. Offloads "help desk" queries from mission staff.
Controls / human review
ATO: Not reported; PIA: Not published