Experiment ·
NAMEDSLOT
Does giving a question-named low-share title one lead passage after the first hits (cut 6 + one searchDetailed rule) raise gold recall without disturbing retrieval breadth, vs cut 6 retrieval replay?
From the record
- What was tested
- NAMEDSLOT
- Result
- recall@5 0.333 → 0.410 but changed 15/60 lists (bar ≤ 12)
- Verdict
- closed: rule failed, CI excluded the hoped-for gain, a step-0 gate failed, or the mechanism was falsified
- Detail
- closed (NOT ADOPTED as frozen: G2c breadth gate FAIL 15/60 vs ≤ 12; G3/G4 not run by stop rule)
- Set
- v0.1 EN 60
- Judge
- none (deterministic gate)
- n
- 60 (v0.1 EN; 39 gold items)
- Seeds
- replay of s44 (deterministic)
- Pre-registered
- yes, sha256-frozen
- Reopen if
- a rule that moves <=12/60 raw lists (generic words resolving to travel guides were the failure)
Linked facts · 2
- R3 Named-slot passage ranking: replay, 0 callsR3 Named-slot passage ranking: replay, 0 callsrecall@5 0.333 → 0.410 but changed 15/60 lists (bar ≤ 12)Kinda record, not a measurement (set, status, judge, n, seeds, date: not recorded)Caveatwhy closed: gate FAILSourceresearch record
reports/CLOSED.md:18Open the data file - Named-slot retrieval (NAMEDSLOT): one lead passage per named title → 15/60 lists changed (bar 12) (gate FAIL)Named-slot retrieval (NAMEDSLOT): one lead passage per named title → 15/60 lists changed (bar 12) (gate FAIL)Kinda record, not a measurement (set, status, judge, n, seeds, date: not recorded)Sourceresearch record
reports/CLOSED.md:18Open the data file