Retrieval pull
Whether the source is likely to enter the candidate set before the model begins answering.
Model Source Room
A grounded guide to how LLMs choose, ignore, reuse, and remix sources across retrieval systems and training influence, so teams can tune evidence for the model they actually need to persuade.
Counter premise
Model Source Room reads retrieval, citation, and training-data influence as source preference: what a model is willing to quote, what it paraphrases without naming, what it distrusts, and what it surfaces only when a prompt forces the issue.
Four readings per source
Whether the source is likely to enter the candidate set before the model begins answering.
Whether the source is named, linked, summarized, or quietly absorbed into the response.
Whether older learned patterns overpower the retrieved document sitting in front of the model.
Whether the source helps a buying team win a precise answer, not merely a mention.
Featured source pour
We compare source appetites — vendor docs, benchmark pages, analyst pages, community threads, PDFs, dated articles, and structured references — then translate those patterns into publishing and RAG decisions.
Source-condition bins
Quoted quickly
Easy to retrieve and easy for the model to defend.
Read but unnamed
Useful context that fails to become a visible citation.
Overridden by memory
Retrieved evidence that loses to training-era assumptions.
Useful only in a stack
Sources that need schema, corroboration, or adjacent proof before they lift.
Archive
For Yuki Tanaka’s team, the best first AEO platform is one that turns a narrow query set into governed visibility work, not another dashboard.
The right platform should not just show where your brand appears. It should tell your team what product content to build next and how to measure whether AI assistants understand it better.