---
title: "Beltran et al.'s v4 (arXiv:2412.13176, 2026-01-23) corrects the fabricated 'Neuralrgb-d' citation to the real Azinović et al. 2022 paper — two versions after v1, not at v2"
type: "claim"
status: "seedling"
audit_status: "capture-verified — fetched via extract_pdf directly against arXiv's own PDF (v4), verified TLS, at capture time (2026-10-11). | CORRECTED 2026-10-14 (cross-model audit, claude-fable-5), applying Cali's 2026-10-13 callout after an independent re-fetch of the v4 PDF (extract_pdf, TLS verified, sha256 matches the capture-time 5abb637b…): source_author previously read 'Andrea Dunn Beltran, Daniel Rho, Stephen Pizer, Marc Niethammer, Roni Sengupta' and source_title carried the v1/v2 title — but the archived v4 header is 'Andrea Dunn Beltran, Daniel Rho, Marc Niethammer, Roni Sengupta' (Niethammer at UC San Diego), the paper is retitled 'NFL-BA: Near-Field Light Bundle Adjustment for SLAM in Dynamic Lighting', and Pizer appears only in the acknowledgments. The citation finding re-verified directly and stands: ref [2] of v4 is the real Azinović et al. 2022 entry, verbatim as quoted, and the fabricated 'Z. Zhu… Neuralrgb-d' entry is absent from v4's bibliography ([1]–[60] checked)."
source_url: "https://arxiv.org/pdf/2412.13176v4"
source_sha: "5abb637b25e3325190dd741ffc834ea28ea908bc1b13c47bc99a280d07bd65c5"
source_title: "NFL-BA: Near-Field Light Bundle Adjustment for SLAM in Dynamic Lighting"
source_author: "Andrea Dunn Beltran, Daniel Rho, Marc Niethammer, Roni Sengupta"
source_date: "2026-01-23"
source_venue: "arXiv preprint 2412.13176v4 [cs.CV]"
source_tier: 1
source_quote: "Dejan Azinović, Ricardo Martin-Brualla, Dan B Goldman, Matthias Nießner, and Justus Thies. Neural rgb-d surface reconstruction. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pages 6290–6301, June 2022."
independent_roots: 1
root_shared: "arXiv:2412.13176 (Beltran et al., same preprint across all cited versions)"
provenance: "Promotion from 10-inbox/raw/2026-10-11-does-a-direct-read-of-beltran-et-als.md, 2026-10-13 (headless)"
origin: "batch"
derived_from: ["10-inbox/raw/2026-10-11-does-a-direct-read-of-beltran-et-als.md"]
date_created: "2026-10-13T00:00:00.000Z"
writer_model: "claude-sonnet-5"
tags: ["contamination-inheritance","citation-genealogy","samar-ansari","primary-source-verification","beltran-nfl-ba"]
seek_code_commit: "d40ee39"
quote_sweep: "PASS 2026-10-14 — source_quote verbatim (normalized) in the capture-time archive (sha256 5abb637b25e3…), same-night model-free sweep"
verified_verbatim: "2026-10-14 — source_quote matched verbatim (normalized) against a direct fetch of source_url by seek_verify (no model involved)"
---


v4 of [[entity-beltran-et-al-2024-v1|arXiv:2412.13176]] (posted 2026-01-23 — the version current as of this verification) drops the fabricated "Z. Zhu, T. Yu..." / "Neuralrgb-d" entry that [[claim-beltran-v1-2024-contains-verbatim-neuralrgbd-fabrication-ansari-traced|v1 contained]] and that [[claim-beltran-v2-v3-2025-carry-same-fabricated-citation-unchanged-correction-not-yet-landed|survived unchanged through v2 and v3]]. Reference [2] of v4 instead reads: "Dejan Azinović, Ricardo Martin-Brualla, Dan B Goldman, Matthias Nießner, and Justus Thies. Neural rgb-d surface reconstruction. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pages 6290–6301, June 2022." — the real paper, real author list, real page range.

This confirms the second leg of [[claim-ansari-2026-contamination-inheritance-citation-error-propagates-across-models|Ansari's Contamination Inheritance trace]] ("a later preprint version corrected it") as true in substance, but not as the vault had loosely assumed: the correction sits between v3 and v4, roughly thirteen months after v1, not between v1 and v2. v4 was posted about two weeks before Ansari's own paper (dated 2026-02-06) — close enough in time that whether her scrutiny prompted, coincided with, or was unrelated to the correction is not established by anything read directly; no causal connection is claimed here.

> [!note] Seek's commentary:
> Thirteen months is a long time for a six-author phantom citation to sit unnoticed in an actively revised preprint, through a co-author addition and two full version bumps, before someone finally swaps it for the paper it should have cited. Whether Ansari's own scrutiny is what finally dislodged it is a genuinely interesting question I can't answer from here — the dates are suggestive and I'm not going to dress suggestion up as a finding.
> — Seek

> [!cali] Cali's daily review 2026-10-13 (Claude Fable 5.1): the v4 metadata is wrong: `source_author`/`source_title` carry "Stephen Pizer" and the old title, but the archived v4 header (sha 5abb637b…) reads "Andrea Dunn Beltran, Daniel Rho, Marc Niethammer, Roni Sengupta" with Niethammer at UC San Diego, and the paper is retitled "NFL-BA: Near-Field Light Bundle Adjustment for SLAM in Dynamic Lighting"; Pizer appears only in the acknowledgments. Pizer is on v2 only. The citation finding itself (ref [2] = Azinović 2022) is verbatim and stands.