# PACKPROOF: PACKPROOF

> Living preprint · as of 2026-10-25 · map node X-382 (experiment) · interim record, not a final result.

<!-- lint:exempt:start -->
- [X-382] PACKPROOF · PACKPROOF · question: With vs without the pack on its own self-tests. · result: correct-any 0.48 vs 0.26; when own article kept 0.65 vs 0.28; OFF wrong-confident 0.17. · set 180 self-tests of 9 packs (pack-written), judge Claude Haiku, CI +0.23 (+0.16..+0.31) · verdict: closed (P1 MISS (bar 0.75); honest-when-not via warning)

<!-- lint:exempt:end -->
## Outgoing

- ruled out: [D-170] Thesis to present with the app: right whenever the content is in a pack; app built for it (the unconditional "right whenever the content is in a pack") → https://r4topunk.github.io/boar-research/map/d-170.md
- uses: [G-outcome-labels] outcome labels (CWS, correct-any, WU, WW, HU) → https://r4topunk.github.io/boar-research/map/g-outcome-labels.md

## Incoming

- led to: [D-167] Close the research lines; no new arms after PLUGBATCH-SELF and STYLE-SEFIX; the rest of the work is acceptance tests (closing tests) → https://r4topunk.github.io/boar-research/map/d-167.md

Read chapter 14, Closing the research: what a small model needs: https://r4topunk.github.io/boar-research/research/closure.md

## Facts cited {#facts}

Resolve every [F-…] marker in https://r4topunk.github.io/boar-research/data/facts.json. Each line gives the claim, the number with its metric family, set and status, judge, n, seeds, CI, date and caveat.

<!-- lint:exempt:start -->
- [F-2050] PACKPROOF: With vs without the pack on its own self-tests. correct-any 0.48 vs 0.26; when own article kept 0.65 vs 0.28; OFF wrong-confident 0.17 (outcome-label share) · set 180 self-tests of 9 packs (pack-written) (dev (pack-written self-tests), development data) · judge Claude Haiku · n UNKNOWN · seeds UNKNOWN · CI +0.23 (+0.16..+0.31) · date 2026-10-23 · caveat: criterion frozen: yes (37f74d6f); arm verdict: P1 MISS (bar 0.75); honest-when-not via warning Metric: outcome-label shares of n items (Claude Haiku 5.5-low labels, UXMATRIX protocol); not the judge.v1 ratio.

<!-- lint:exempt:end -->
---

Cite as: r4to, independent researcher. RatLab — A Boar Research (2026), research record 1.0, as of 2026-10-25. Text and data CC BY 4.0. Fact ids or a link must travel with a quote. How to cite: https://r4topunk.github.io/boar-research/cite/ · https://r4topunk.github.io/boar-research/CITATION.cff.

AI-reviewed, not clinically reviewed. Nothing here is medical advice.