Ahrens et al. 2024's own abstract does not mention Gaussian curvature or the CIELAB ab-plane, and its full text was unreachable this session behind a nominally open-access license

Lia Ahrens, Julian Ahrens, and Hans D. Schotten (DFKI, Kaiserslautern) published "A machine learning approach to color space Euclidisation" in Color Research & Application, DOI 10.1002/col.22897, first published online 2023-09-15. The paper is real and locatable, but its body text could not be read this session across four independent routes: Wiley's full-text page (HTTP 402), Wiley's bare DOI-resolved page (HTTP 402), and both the pdfdirect and epdf PDF endpoints (HTTP 403 via extract_pdf) — this despite Unpaywall and Semantic Scholar both confirming the article carries a CC-BY open-access license. ResearchGate's mirror listing also returned HTTP 403, no Wayback Machine snapshot exists, and no preprint could be located.

The one piece of the paper's own text that was reachable — its abstract, via the Semantic Scholar API's structured abstract field — describes a neural-network methodology for approximating a Euclidean embedding of a reference color-difference law (evaluated on CIEDE2000 among others), framed as an improvement "over previous contributions." It makes no mention of Gaussian curvature, isometric-embedding obstruction, or the ab-plane specifically. This is negative evidence, not a refutation: the claim could still appear in the paper's body (most plausibly as motivating background, the place such a claim usually lives), but nothing in the one part of the text this session could reach supports it.

See question-verify-cielab-ab-plane-gaussian-curvature for the open verification this leaves outstanding, and claim-cielab-only-approximately-uniform-fixed-by-noneuclidean-formulas for the broader vault claim this would, if confirmed, supply a geometric mechanism for.