Investigation Report: Reference Anomalies in "Attribute-Based Bilateral Access Control Scheme with Policy Hiding"

Academic fraud report · Geng Detector

Summary

This report examines a 2025 online-first paper published in Journal of Shanxi University (Natural Science Edition) by Guo Lifeng, Sun Wenzhe, and Zhao Yu (DOI: 10.13451/j.sxu.ns.2025008), titled "Attribute-Based Bilateral Access Control Scheme with Policy Hiding" (策略隐藏的基于属性的双边访问控制方案). The overall verdict is questionable (🟡). The primary concrete finding concerns reference [13], where the author list reads "RIEPEL D, WEE H, RIEPEL D, et al." — an implausible duplication of the first author as the third author, indicating a careless copy-paste or AI-hallucination error in reference generation that survived peer review. A secondary finding notes that Figures 2–3 and tabular performance data cannot be verified at the pixel or raw-data level because the source PDF lacks accessible numerical detail, preventing standard image-duplication and statistical outlier checks. Confidence is moderate: the reference error is directly verifiable in the text, but image/data integrity remains unassessable. No fabrication or misconduct is proven, only sloppy scholarship.

Verdict

🟡 Questionable. One directly verifiable bibliographic error (duplicate author in reference [13]) survives editorial scrutiny; image and raw-data integrity remain unverified.

Key findings

Reference [13] author duplication (severity 🟡, verified). The bibliography entry lists authors as "RIEPEL D, WEE H, RIEPEL D, et al." — the first and third author names are identical, which is not credible for a real citation. This pattern is typical of careless reference-manager export or AI-generated bibliographies and indicates inadequate proofreading.
Image and raw-data verification limited (severity: undetermined, evidence insufficient). Figures 2 and 3 and the experimental performance tables are referenced in Section 5.2 ("性能对比"), where the authors state experiments were repeated 50 times and averaged, with decryption times reportedly under 1 ms. The PDF as provided does not expose pixel-level image data or the underlying raw timing data, so standard image-duplication detection (e.g., one-figure-reuse) and statistical outlier tests cannot be performed.
No evidence of fabrication, falsification, or plagiarism was identified within the limits of text-only review.

Evidence highlights

Direct textual evidence: reference [13] reads exactly RIEPEL D, WEE H, RIEPEL D, et al. Fabeo[C]… — duplicated surname/initials for author positions 1 and 3.
Section 5.2 of the paper states: "共进行50次实验取平均值" (50 experiments averaged) and reports decryption time "稳定在1ms以内" (stable within 1 ms); these claims are plausible for outsourced-decryption cryptographic schemes but cannot be cross-checked without raw data.
DOI: 10.13451/j.sxu.ns.2025008 — preserved verbatim.
Authors: 郭丽峰 (Guo Lifeng), 孙文哲 (Sun Wenzhe), 赵宇 (Zhao Yu).
Journal: 山西大学学报(自然科学版) / Journal of Shanxi University (Natural Science Edition), 2025 online-first.

Notes

The duplicate-author error, while low-severity on its own, is a marker of insufficient editorial polish and is consistent with automated bibliography generation; it warrants an erratum or full reference-list audit.
Because cryptographic papers rely on reported performance numbers to support claims of efficiency, the inability to verify the 50-run averaged timings and the figures is a notable limitation of this review; a request for raw experimental logs from the authors would resolve this.
All findings are based on the supplied PDF text; no peer-review correspondence or author raw data were available.
This report is AI-assisted and intended for academic discussion; final determinations of misconduct require investigation by qualified institutions.