Video, , 9:43
Extraction Arena: Evaluate Vision LLMs for Document Extraction
A walkthrough of the harness behind the note.
Summary
A recorded walkthrough of Extraction Arena, the open-source harness I use to evaluate vision LLMs on document extraction, field by field.
It is the same argument as the Extraction Arena note, recorded: three vision models, one gold rescue sheet, and a score that moves when the question changes. The code is on GitHub.
Play video (opens YouTube in a new tab)
// the player loads from YouTube only when you press play
Also on YouTube, on my channel @martin-cousseau.