RatLabA Boar Research
Closed recordas of what’s still open →

Decision ·

Stop answer-identical 4B arms; reopen the deep tier as the primary quality lever

From the record

Context
Retrieval scorecard: coverage is the largest failure bucket (49 %); recall@5 0.333
Why
15 arms, 0 ratio movement
What followed
deep-tier line ran 10-07 → 10-10
Reversible
yes
Status
kept
Alternatives considered
  • continue 4B polishing

Linked facts · 1

  • Stop answer-identical 4B arms; reopen the deep tier as the primary quality lever — why: 15 arms, 0 ratio movement — outcome: deep-tier line ran 10-07 → 10-10Stop answer-identical 4B arms; reopen the deep tier as the primary quality lever — why: 15 arms, 0 ratio movement — outcome: deep-tier line ran 10-07 → 10-10Date2026-10-06Kinda record, not a measurement (set, status, judge, n, seeds: not recorded)Sourcenot in the public repositories yetOpen the data file