Differences

This shows you the differences between two versions of the page.

--- provenance:statistics:annotation [2026/09/11 03:17] – Add the fifth review pass, which found that two independently-accepted fixes composed into a false claim, and record the DokuWiki heading-markup defect the run caught itself. Regenerated from the current script output. Authored by Claude karel.kubicek.claude
+++ provenance:statistics:annotation [2026/09/17 07:25] (current) – Markup rendering sweep: wrap escapes, CLI flag typography, and heading quotes. Authored by Claude. karel.kubicek.claude
@@ Line 18: / Line 18: @@
 | Who | Claude (Opus 5) wrote the scripts, ran the queries, read the audit candidates and drafted the page. Review layer below. |
 | New scripts | ''scripts/gt_fold.mjs'', ''scripts/annotation_audit.mjs'', ''scripts/report_annotation.mjs'', ''pages/holdout_by_site.py'' |
-| Modified | ''scripts/quote_check.mjs'' gained a ''--method'' selector (additive; no existing selector changed) |
+| Modified | ''scripts/quote_check.mjs'' gained a ''%%--method%%'' selector (additive; no existing selector changed) |
 | Bibliography | 2 entries added (''arp2022_dosdonts'', ''pendlebury2019_tesseract''). ''xiong2026_tgnn'' was **already present** and was not re-added — see [[#Bibliography]] |
@@ Line 175: / Line 175: @@
 ===== Quotes checked =====
-''node scripts/quote_check.mjs --method llm'' — all **255** evidence quotes on the LLM classification **tuples** carried by the **177** papers, against the rendering the extractor actually read, with a fallback to a second rendering of the PDF. (255 is a tuple count, not a paper count; an earlier draft of this line wrote "the 177 LLM classification tuples" and attached the wrong noun's number.)
+''node scripts/quote_check.mjs %%--method%% llm'' — all **255** evidence quotes on the LLM classification **tuples** carried by the **177** papers, against the rendering the extractor actually read, with a fallback to a second rendering of the PDF. (255 is a tuple count, not a paper count; an earlier draft of this line wrote "the 177 LLM classification tuples" and attached the wrong noun's number.)
 ^ Verdict ^ Quotes ^
@@ Line 1448: / Line 1448: @@
 ===== Fold self-tests and the FULL unmapped residue =====
-''node scripts/gt_fold.mjs --residue''. Every one of the 2,206 distinct unmapped strings with its tuple count, so the part the fold could not classify stays visible.
+''node scripts/gt_fold.mjs %%--residue%%''. Every one of the 2,206 distinct unmapped strings with its tuple count, so the part the fold could not classify stays visible.
 <file text gt_fold-residue.txt>
@@ Line 4062: / Line 4062: @@
 ===== Quote check (unedited) =====
-''node scripts/quote_check.mjs --method llm'' — all 255 evidence quotes on the 177 LLM classification tuples. The 18 below threshold in both renderings are listed at the end and were read by hand; see [[#Quotes checked]].
+''node scripts/quote_check.mjs %%--method%% llm'' — all 255 evidence quotes on the 177 LLM classification tuples. The 18 below threshold in both renderings are listed at the end and were read by hand; see [[#Quotes checked]].
 <file text annotation_quotecheck-output.txt>
@@ Line 4766: / Line 4766: @@
 <bibtex bibliography></bibtex>
+===== Markup sweep, 2026-09-17 =====
+Mechanical rendering repair only: a fresh live raw/XHTML export of 188 pages was checked with ''check_wrap.mjs'' and ''check_typography.mjs''. Affected plugin tags, CLI flags and heading markup were repaired; no figures or substantive prose were changed. The resulting source and rendered DOM were re-checked after saving.