Chatting with Papers: A Hybrid Approach Using LLMs and Knowledge Graphs

This demo paper reports on a new workflow $\mathit{\text{GhostWriter}}$ that combines the use of Large Language Models and Knowledge Graphs (semantic artifacts) to support navigation through collections. Situated in the research area of Retrieval Augmented Generation, this specific workflow represents the creation of local and adaptable chatbots. Based on the tool-suite $\mathit{\text{EverythingData}}$ at the backend, $\mathit{\text{GhostWriter}}$ provides an interface that enables querying and ``chatting’’ with a collection. Applied iteratively, the workflow supports the information needs of researchers when interacting with a collection of papers, whether it be to gain an overview, to learn more about a specific concept and its context, and helps the researcher ultimately to refine their research question in a controlled way. We demonstrate the workflow for a collection of articles from the $\mathit{\text{method}}\mathit{\text{data}}\mathit{\text{analysis}}$ journal published by GESIS – Leibniz-Institute for the Social Sciences. We also point to further application areas.