Academic fraud detection report on 'The Battery Drain Puzzle: Component-Level Modeling of Smartphone Battery Drain for SOC'

Academic fraud report · Geng Detector

Summary

This report examines a manuscript (DOI: 10.3390/1010000) submitted to a presumed MDPI journal, dated March 25, 2026. The verdict is confirmed academic misconduct (real evidence). Rather than original research, the paper appears to be a mathematics modeling competition entry (likely MCM/ICM or a national contest) repurposed as a journal submission without meaningful adaptation. Key issues include: entirely unaltered placeholder author names (Firstname Lastname, e-mail@e-mail.com, Affiliation 1); verbatim inclusion of competition-task prompt language such as 'we need to address the following tasks'; structural hallmarks of a contest paper (Assumptions and Justifications, Notations, Model Strengths and Weaknesses sections); broken figure/table references like 'As shown in Figure.'; unresolved citation '[?]'; a placeholder DOI (10.3390/1010000); and tables with suspiciously neat train/test splits. Confidence is high for the template/placeholder evidence; lower for the contest-source allegation, which requires comparison with known contest archives. Pixel-level image analysis was not possible. No precise numbers beyond the listed TTE comparisons (576 vs 564, 247 vs 246, 178 vs 178) are asserted.

Verdict

⛔ Confirmed academic misconduct (real evidence). The manuscript is not an original research article but a near-verbatim repackaging of a mathematics modeling competition paper (likely MCM/ICM or a Chinese national contest) inserted into a journal template without basic remediation.

Key findings

Unedited template placeholders: Author names appear as Firstname Lastname, contact as e-mail@e-mail.com, affiliation as Affiliation 1; copyright footer reads Submitted to possible open access publication under the terms and conditions of the Creative Commons Attribution (CC BY) license.
Competition prompt text retained: Introduction (~line 30) contains Based on the problem background and constraints outlined, we need to address the following tasks:, an instruction-to-solver phrasing absent from genuine journal articles.
Resolvable artifacts: unresolved citation [?] (line 217); missing author/title in Reference 5; empty Reference 11; concatenated figure tokens such as Figure 7.Figure 8. and As shown in Figure. with no number.
Contest-paper structure: dedicated Assumptions and Justifications, Notations, and Model Strengths and Weaknesses sections match MCM/ICM scoring conventions, not standard IMRaD article design.
Placeholder DOI: 10.3390/1010000 is consistent with an auto-generated MDPI placeholder, not an assigned identifier.
Data presentation anomalies: Table 2 reports TTE prediction-vs-actual pairs of 576 vs 564, 247 vs 246, and 178 vs 178; the near-perfect agreement is consistent with either a problem-set dataset or simulated output rather than original smartphone experiments.
No substantive experiments: No materials-and-methods section capable of supporting real-device battery-drain measurements; submission timeline (2026-03-25) and latest cited 2024 Nature Energy reference are internally coherent but do not corroborate authorship.

Evidence highlights

DOI: 10.3390/1010000 (placeholder).
Submission metadata: Received / Submitted: March 25, 2026.
Quoted lines preserved exactly: Based on the problem background and constraints outlined, we need to address the following tasks:; • substitute for the model).; Translate model findings into practical user recommendations...; As shown in Figure.; [?].
Reference anomalies: Reference 5 author/title missing; Reference 11 empty.
Table 2 values: 576 vs 564, 247 vs 246, 178 vs 178 (TTE prediction vs actual).

Notes

Pixel-level image manipulation analysis (duplication, splicing, reuse) was not feasible because source image files were not supplied; all conclusions here are text-based and structural.
The strongest, fact-grade claims concern template placeholders and verbatim prompt language; the allegation that the underlying content is a specific MCM/ICM paper is plausible but requires external matching against contest archives before being treated as definitive.
Indicators of misconduct are documented; final adjudication remains with the institution or publisher.