Scoring & evidence
Human baseline
Performance achieved by an appropriate group of people on the same task.
Definition
A human baseline measures how trained users or subject-matter experts perform under comparable instructions, time and information conditions.
Why it matters
It gives a score practical context and helps avoid treating imperfect reference answers as absolute truth.
Related concepts
- Golden set
A curated set of representative inputs and trusted expected outcomes.
- Rubric
Explicit criteria describing what a good answer or deliverable must contain.
- Inter-judge agreement
How consistently two or more graders assess the same outputs.