Marcel Bautista-Kuljevan commited on
Commit
c9fc206
·
unverified ·
1 Parent(s): b57ce70

Simplify explanation of AI benchmarking

Browse files

Removed redundant text about AI transcription.

Files changed (1) hide show
  1. README.md +1 -2
README.md CHANGED
@@ -31,8 +31,7 @@ compares each output to the ground truth at every relevant level (text,
31
  ALTO, PAGE, entities, reading order), and produces a self-contained HTML
32
  report with factual numbers, statistical tests and a reproducibility
33
  snapshot. Without ground truth, no benchmark — Picarones measures how
34
- well an AI matches a known reference, not how well it transcribes an
35
- arbitrary document.
36
 
37
  Typical workflow: assemble a small golden dataset whose profile (script
38
  type, period, language) matches the production corpus you intend to
 
31
  ALTO, PAGE, entities, reading order), and produces a self-contained HTML
32
  report with factual numbers, statistical tests and a reproducibility
33
  snapshot. Without ground truth, no benchmark — Picarones measures how
34
+ well an AI matches a known reference.
 
35
 
36
  Typical workflow: assemble a small golden dataset whose profile (script
37
  type, period, language) matches the production corpus you intend to