Lee Sharks
A reproducible instrument for measuring how faithfully AI summarization systems convey scholarly knowledge.
There is a lack of objective methods to evaluate the fidelity of information when AI-based summarization systems summarize scholarly literature.
Construct two graphs (scholarly graph and retrieval graph) and quantify their differences using variables such as compression loss, invention, and distortion. Track changes over time via DOI-anchored temporal snapshots.
Provides a reproducible measurement protocol and coding rubric, including a worked example using Revelation studies. This contributes to evaluating the reliability of AI summarization systems.