📏Benchmark Saturation
Also known as Saturated Benchmark
Top models all score near full marks. So the test can no longer tell them apart.
Think of it like
It is like a high jump bar set too low. Everyone clears it, so it shows no winner.
Example
Three models score 91%, 92% and 92.5% on an old quiz. Jack cannot tell which is best, so he looks at a newer, harder test.
Why it matters
A saturated test makes every new model look the same. Teams need to know which tests still have room to show a difference.
Definition
Common mistake
Key takeaways
- 1Ceiling reached. Scores bunch up near full marks.
- 2Noise wins. Small gaps mean little.
- 3Harder tests next. New tests replace the old.