Canonical documentation for RunLog Atlas
Understanding the fundamental distinction
How confidence scores work
How human review works
How corrections propagate
How multiple reviewers work together
System architecture overview
Key terms and definitions