[clinic · EX-05] Make starvation reproduce

[akash-coded]

Exercise clinic. One long-running thread per exercise. Ask here rather than opening a
new thread — the answers accumulate where the next person will look for them.

Exercise: EX-05 — Make starvation reproduce · difficulty 🔴

What this exercise asks

Read the exercise first; this thread is for when you are stuck, not instead of it.

Three ways people get stuck on this one

1. Skewing the document count but not the entity prevalence. Starvation is driven by the ratio between the two compared entities, not by corpus size.

2. Measuring aggregate recall. Starvation is confined to comparison questions and averages away — the same failure as #1.

3. Expecting it at a ratio near 1. The README says it does not reproduce at ≈1 by construction; you have to make the ratio large.

How to ask here

Post the numbers you got, the config that produced them, and what you already tried.
A question with a measurement in it gets answered in one round trip; a question without one
takes three.

If you have finished it, post the write-up in Show and Tell with the
[solution · EX-05] prefix rather than here — this thread stays for people who are stuck.