# HEADPASS-X2b: HEADPASS-X2b: neighbour chunks + budgeted dense candidates appended after the keep step

> Living preprint · as of 2026-10-25 · map node X-161 (experiment) · interim record, not a final result.

<!-- lint:exempt:start -->
- [X-161] HEADPASS-X2b · HEADPASS-X2b: neighbour chunks + budgeted dense candidates appended after the keep step · question: Does appending the retrieved article's neighbouring chunks plus up to 2 dense-vector candidates (≤ +600 prompt tokens) raise the 4B's score on the 180 pack-matched self-tests? · result: Opus +0.066 (+0.043…+0.091); Haiku +0.056; fact chunk present in context 35 → 102 of 180. · set 180 pack self-tests (dev), judge Claude Opus (L2) and Claude Haiku (L1), CI +0.043…+0.091 · verdict: adopted

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

- uses: [G-x2b] HEADPASS-X2b → https://r4topunk.github.io/boar-research/map/g-x2b.md

## Incoming

- led to: [D-078] Pivot to Knowledge Packs; primary metric = in-pack conditional ratio (the pack line's one lever) → https://r4topunk.github.io/boar-research/map/d-078.md
- ruled out: [X-164] HEADPASS-X3: Cross-encoder reranker choosing the appended chunks on the HEADPASS path (a reranker added nothing over X2b) → https://r4topunk.github.io/boar-research/map/x-164.md

Read chapter 9, Knowledge packs: https://r4topunk.github.io/boar-research/research/packs.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-028] The one retrieval change that converted: append the retrieved article's neighbouring chunks plus a few vector-search candidates, under a 600-token budget (HEADPASS-X2b). Opus +0.066 (+0.043…+0.091); Haiku +0.056; fact chunk present in context 35 → 102 of 180 · set 180 pack self-tests (dev (designed on the same set), development data) · judge Claude Opus (judge.v1, both orders) · n 180 · seeds 1 · CI paired 95 % bootstrap · date 2026-10-14 · caveat: LEVER on pack-matched dev data only; inert on the claim kit (ceiling ≈ 0.011) and not a bounty-bar lever.
- [F-1069] HEADPASS-X2b (X1 + budgeted dense candidates, +600 tokens per item) fact present 102/180; Opus +0.066; Haiku +0.056 · set the 180 dev (dev, development data) · judge Opus + Haiku · n 180 · seeds 1 · CI (+0.043…+0.091) / (+0.036…+0.076); n 180; 1 seed · date 2026-10-14 15:30 · caveat: LEVER by frozen rule on pack-matched dev only; designed on the same 180

<!-- 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.