Review date: 10 September 2026. Primary reviewer: independent source compilation reviewer. Frozen state: the repository as it stood on 2026-09-10T05:48:43Z. The [[diophantine_problems/pipeline_math_2026_tiling_complement/evidence/verify/compilation_review|compilation review]] retains the verdict and independent derivations. This is a rendition of the original coverage receipt, not new source reading during filing.

Native subjects

The following aliases expand to repository-relative canonical homes:

• P = library/diophantine_problems/pipeline_math_2026_tiling_complement
• C = library/diophantine_problems/corvaja_zannier_2011_abcd_function_fields
• H = library/diophantine_problems/heath_brown_2009_sums_differences_three_kth_powers

These are exactly nine Markdown subjects, read as they stood on 2026-09-10T05:48:43Z.

The frozen consumer was read only for formulation and scope: wiki/problems/diophantine_problems/E0477/_index.md as it stood on 2026-09-10T05:48:43Z. Its variants and status-search claims are not review subjects.

The supplied scope note was titled NEXT_SOURCE_ACCOUNT_TRIAGE.md (working storage; not retained). It is assignment context, not a mathematical premise or required native artifact. It pre-directed coordinate normalization, the shell repair and the family convention. Two reported weakest-step selections were therefore pre-identified; the reviewer independently derived their checks, not their selection.

Exposure ruling. Besides the unretained triage note that assessed the reconstruction as complete and pre-named two of the three weakest steps, the commissioned read set as it stood on 2026-09-10T05:48:43Z carried the subject pages' own author-recorded standing (under library/diophantine_problems/pipeline_math_2026_tiling_complement/: _index.md lines 171–179, lemma_1_4.md 239–244, corollary_1_5.md 58–64, proposition_1_6.md 229–233, lemma_1_7.md 96–101, proposition_1_8.md 70–77, theorem_1_1.md 116–122) and public-acceptance text for the existence conclusion (_index.md lines 56–75; wiki/problems/diophantine_problems/E0477/_index.md line 7, lines 21–26 and 44–57); a separately commissioned materiality grader acting in the corpus grader role (model: Claude Fable 5.1) ruled on 18 September 2026 by the content test that all three exposures are immaterial, because none states or implies the premise-relative verdict asked of the reviewer and the review's reasoning rests on independent rederivation corroborated by the subreview and the grader, so the pass with named corrections stands unchanged.

Sources read and complete-page reading

All four PDF hashes agree with the supplied triage. No source was fetched.

There were no visual or access gaps on these selected pages. Rendered formulas were legible. Extraction corrupts some signs, membership symbols, and other glyphs, particularly in the manuscript; the rendered images controlled the mathematical reading. No conclusion rests on extracted formulas alone.

The unselected Corvaja-Zannier printed pp. 439–453 and unselected Heath-Brown pages were not read by the primary reviewer. The original Mason-Stothers and Brownawell-Masser proofs, the original sources inside Heath-Brown's determinant proof, historical Erdős-Graham pages, Price's manuscript, and stronger exponent/input variants were not opened. Unrelated external main theorems receive no review credit.

Source/result correspondence and checked local additions

Supplementary independent interface reading

A separate fresh-context subreview read the frozen H/theorem_2.md and P/proposition_1_6.md, visually inspected journal pp. 1579-1580 and 1589 and v1 pp. 1–2 and 10-11, checked their extracted text, and verified the corresponding four hashes before and after reading. It returned refutation-failed for the source interface and fixed-$c$ shell-to-box deduction. It also checked the surrounding signed witness bound, nonsingularity, projection, and exponent while assuming Corollary 1.5. It did not review that corollary or the external determinant proof.

The primary review's derivation and verdict were reached and communicated before this subreview returned. Its findings agreed with the primary review and were compared only afterward. Seven supplementary renders were made, covering the listed journal and v1 pages. They are disposable reading aids, not required native artifacts. The subreview wrote no separate report or corpus file, fetched no source and ran no mathematical computation. This record retains the primary reviewer's attributed account of that supplementary coverage; it does not invent a separate retained subreview.

The four editions read are identified on their source cards: the manuscript, Corvaja-Zannier, Heath-Brown journal, and Heath-Brown v1. The links identify editions, not an instruction to fetch or substitute a different version.

Reproducible source inspection

System Poppler tools pdfinfo, pdftotext -layout and pdftoppm were used. The 17 primary and seven supplementary rendered images are disposable, regenerable reading aids. Their private locations are not mathematical dependencies, and the native filing does not require those old pixels.

The following commands reproduce the primary reading scope from the repository root with copies of the four PDF editions named above at the paths assigned below; the reader supplies the copies. They create only a new disposable rendering directory and were not run during native filing:

reading_output="$(mktemp -d)"
pipeline_pdf=library/diophantine_problems/pipeline_math_2026_tiling_complement/pipeline_math_2026_tiling_complement.pdf
cz_pdf=library/diophantine_problems/corvaja_zannier_2011_abcd_function_fields/corvaja_zannier_2011_abcd_function_fields.pdf
hb_journal_pdf=library/diophantine_problems/heath_brown_2009_sums_differences_three_kth_powers/heath_brown_2009_sums_differences_three_kth_powers.pdf
hb_v1_pdf=library/diophantine_problems/heath_brown_2009_sums_differences_three_kth_powers/heath_brown_2009_sums_differences_three_kth_powers_arxiv_v1.pdf
pdftoppm -r 110 -f 1 -l 6 -png "$pipeline_pdf" "$reading_output/pipeline"
pdftoppm -r 125 -f 1 -l 3 -png "$cz_pdf" "$reading_output/cz"
pdftoppm -r 125 -f 19 -l 19 -png "$cz_pdf" "$reading_output/cz"
pdftoppm -r 110 -f 1 -l 2 -png "$hb_journal_pdf" "$reading_output/hbj"
pdftoppm -r 110 -f 11 -l 11 -png "$hb_journal_pdf" "$reading_output/hbj"
pdftoppm -r 110 -f 1 -l 2 -png "$hb_v1_pdf" "$reading_output/hbv1"
pdftoppm -r 110 -f 10 -l 11 -png "$hb_v1_pdf" "$reading_output/hbv1"

For each of those four assigned paths, pdfinfo <source.pdf> identifies the document, pdftotext -layout <source.pdf> - extracts all text, and pdftotext -f <first> -l <last> -layout <source.pdf> - extracts a selected PDF-page range. Substitute one assigned path and the page numbers listed above. Every emitted selected page was opened during the original reading. There is no mathematical checker or runtime-dependent certificate to reproduce for this noncomputational argument.

Grader-supplied source context and limits

The distinct grade reports checking the native subject hashes, all four PDF identities and the substantive mathematics against those sources. Its findings add same-page support for the family convention: journal p. 1580, Theorem 1, calls the $d=1$ case “essentially different linear parameterizations.” This support is attributed to the grader, not to a new primary-reviewer finding or to the native filing author. It leaves the introductory definition's unprinted nonconstancy qualification explicit.

The original coverage receipt remains unchanged in working storage. This native rendition keeps every subject, version, selected page, exclusion, result correspondence and supplementary scope, while removing private paths and making rendering aids disposable. It is not a new reading.

Before the original native filing, all ten Markdown hashes and four actual PDF hashes again matched the recorded values. The frozen native subjects are named by path and date, 2026-09-10T05:48:43Z; the text they carried then is not retained as copies. The accepted transformation and subsequent standing reconciliation change the seven source pages' verification wording and generated navigation, not their statements or proofs. The PDF digests this record carried are not restated. Both external interfaces, the four editions read and the consumer remain unchanged.

The accepted native transformation and same-checkpoint standing reconciliation record the seven source pages' independent compilation review within the exact premise-relative scope identified by the review and grade. Attack selection was partly pre-directed; derivations were independently performed. External proofs, catalog status, public acceptance, other exponents or input variants, effectiveness and formal verification receive no additional coverage.