Opening Pandora's box: Paper mills in conference proceedings

Abstract: Paper mills are a growing threat to the integrity of science, yet their penetration in conference proceedings remains underexplored despite conferences being more important than journals in some scientific subfields. This study aims to identify papers in conference proceedings whose titles have been offered for sale on social media platforms. We collected more than 4,000 unique publication offers from more than 200 social media channels and used semi-automated methods along with human assessment to match offers with papers published in IEEE conference proceedings. We identified 1,720 papers in 286 IEEE conference proceedings, accounting for up to 23.51% of an individual conference. These problematic papers are co-authored by more than 6,500 researchers from over 3,500 affiliations in 55 countries. The identified papers demonstrate collaboration anomalies, high diversity of affiliations per paper, citation manipulation, a predominance of six-author papers, and content-based irregularities. Our findings show that paper mills are a large, organized, and often public market that commercializes scientific misconduct, not limited to papers, but infiltrating multiple parts of the research ecosystem.

• The paper introduces a hybrid detection methodology that identifies paper mills based on fixed co-authorship patterns and textual anomalies.
• The study reveals that up to 23.51% of IEEE conference proceedings may be compromised by authorship-for-sale operations, affecting research metrics.
• It highlights systemic issues such as citation manipulation and distorted evaluation metrics, urging publishers to implement targeted remedial measures.

Systematic Analysis of Paper Mill Penetration in IEEE Conference Proceedings

Introduction

The proliferation of paper mills poses a significant threat to the integrity of academic publishing, especially within conference proceedings where peer review standards and editorial scrutiny frequently lag behind those of journals. The study "Opening Pandora's box: Paper mills in conference proceedings" (2604.22458) delivers a comprehensive, data-driven exploration of authorship-for-sale operations, focusing on IEEE conferences given their centrality to several fields and their demonstrated vulnerability to mass malpractice. Leveraging over 4,000 unique publication offers extracted from more than 200 social media channels, the study introduces semi-automated matching methodologies to trace the explicit connection between authorship-for-sale advertisements and published conference papers.

Empirical Findings: Prevalence and Patterns

The investigation identified 1,720 potentially problematic papers within 286 IEEE conference proceedings, with individual venues showing up to 23.51% infiltration. The analysis establishes several characteristic anomalies distinguishing paper mill products:

• Fixed Co-Authorship Pattern: Over 82% of identified papers feature exactly six authors, contrasting sharply with baseline distributions for IEEE conferences. This fixed slot count aligns with seller offers and conference-specific limits.
• High Institutional Diversity: Papers display an average of 5.1 affiliations per manuscript, with 43% incorporating international co-authors. This is indicative of transactional, rather than organic, collaboration.
• Citation Manipulation: Direct evidence of citation boosting services is observed, including cases of references with no topical relevance and citation cartels targeting metrics optimization.
• Content Irregularities: There is rampant reuse of titles, abstracts, and full passages across papers, frequent presence of tortured phrases (indicative of paraphrasing tools), superficial analyses, and plagiarized content.
• Commercial Email Usage: Non-institutional emails are commonly used, sometimes recurring across unrelated co-authors, suggestive of paper mill facilitation.

Over 6,500 researchers from 3,500 distinct affiliations co-authored these papers, predominantly from India, mirroring the target market for the authorship-for-sale services. Some individuals appeared as co-authors in more than 50 identified papers, further substantiating systemic exploitation.

Methodological Innovations

The study utilizes a hybrid of manual searches (Google Scholar, IEEE Xplore, Scopus) and automated title-matching (Levenshtein distance, Sentence-BERT embeddings) against the corpus of collected offer titles, followed by human verification. This methodological rigor ensures high-confidence identification of paper mill origins, supplemented by auxiliary checks for collaboration, textual, and citation anomalies.

Theoretical and Practical Implications

Integrity Risks and Evaluation Bias

The empirical evidence underscores that paper mills have matured into large-scale, organized markets commercializing research misconduct. Conference proceedings, especially those organized by decentralized third-party entities and local IEEE sections, are particularly susceptible due to lack of publisher oversight and extremely short submission-to-acceptance timelines (10 days or fewer in many cases).

Authorship-for-sale fundamentally distorts academic metrics: the prevalence of transactional collaborations inflates co-authorship statistics, international collaboration indices, and individual citation scores, thus biasing scientometric analyses and compromising the utility of these metrics for evaluation, hiring, and funding decisions.

Citation Manipulation

Citation boosting and reference padding by paper mills undermine the reliability of scientometric measures that are widely used for career advancement and funding allocation. The vulnerability of citation-based research evaluation is further exacerbated by artificial citation networks and cartels.

Systemic Vulnerabilities

The study identifies critical systemic flaws: conferences in IEEE's operational manual are treated as "Other Publications" with minimal quality benchmarks, while Scopus-indexed proceedings count toward institutional evaluation metrics. This mismatch creates incentives for manipulation and low-cost mass publication, directly abetted by paper mills.

Recommendations for Remediation

Three targeted interventions are proposed:

1. Systematic Screening for Red Flags: Publishers should track authorship counts, affiliation diversity, text reuse, and tortured phrases throughout the submission lifecycle, not just at the camera-ready stage.
2. Proactive Social Media Monitoring: Publishers must surveil social platforms for authorship-for-sale and citation-boosting offers, adapting detection protocols as paper mills evolve.
3. Expedited Investigation and Retraction: Develop quick-response protocols for systematic manipulation, informed by COPE guidelines, to supersede piecemeal, paper-by-paper investigations.

Implementation of these measures necessitates scaling research integrity teams and leveraging existing detection tools to match the growing scale of paper mill operations.

Limitations and Scope

The analysis is limited by language and platform bias, favoring English-speaking channels and potentially underestimating penetration in China and other regions. The dataset is also temporally bounded, without full capture of channels deleted or offers renamed by mill operators.

Future Directions

The findings prompt several future research avenues:

• Development of robust, automated screening pipelines for large-scale detection in conference submissions.
• Institutional and publisher-level reforms to publication evaluation criteria, decoupling career incentives from vulnerable metrics.
• Cross-disciplinary studies to track migration of paper mill activity from conferences to other publication types (journals, books, patents).
• International collaboration for standards enforcement and fraud mitigation, particularly among indexers such as Scopus and Web of Science.

Conclusion

This study provides a systematic, quantitative account of paper mill penetration in IEEE conference proceedings, exposing organized, market-driven misconduct with measurable impact on the research ecosystem. The deployment of semi-automated detection and network analysis advances the empirical toolkit available for combating authorship-for-sale. The scale and diversity of the penetration demand urgent remedial action from publishers, indexers, and research institutions to restore confidence in conference proceedings as legitimate venues for scholarly communication.