aiqbench analysis
This research demonstrates how authorship attribution systems can falsely implicate legitimate researchers by mistaking inherited academic stylistic patterns from advisors or shared mentors for AI generation. The findings highlight a critical vulnerability in current detection tools, showing they frequently flag true authors who share an advisor with the accused as potential ghostwriters rather than identifying actual misconduct. For the community, this underscores the urgent need to refine models that distinguish between genuine generative assistance and natural mentorship influences on writing style.
Writing Style Similarity Reflects Academic Genealogy
arXiv:2608.14843v1 Announce Type: new Abstract: As authorship attribution systems are increasingly deployed to detect ghostwritten and AI-generated papers, their errors can support accusations against legitimate authors. These systems assume each author's style is their own. Researchers, however, study under advisors, and inherit their stylistic quirks. We build a corpus of arXiv authors with $\geq 2$ solo papers from the Mathematics Genealogy Project graph, giving $5{,}803$ total authors and $2{,}501$ ground-truth advisor-student pairings. Using embeddings from a fine-tuned model, advisors sit $39.9\%$ closer in cosine distance to their students than a random same-field author does. Two open encoders reproduce the effect at $12.6\%$ and $14.5\%$. \emph{Academic siblings}, two students of one advisor who may never have met, sit $30.4\%$ closer across $8{,}360$ pairs, even when they studied at different institutions. Pairs who share only an institution and a field show negligible similarity. Given a closed-set attribution task over the same corpus, the system's errors occur on the true author's advisors and academic siblings $11$ times more often than chance.
Abstract:As authorship attribution systems are increasingly deployed to detect ghostwritten and AI-generated papers, their errors can support accusations against legitimate authors. These systems assume each author's style is their own. Researchers, however, study under advisors, and inherit their stylistic quirks. We build a corpus of arXiv authors with $\geq 2$ solo papers from the Mathematics Genealogy Project graph, giving $5{,}803$ total authors and $2{,}501$ ground-truth advisor-student pairings. Using embeddings from a fine-tuned model, advisors sit $39.9\%$ closer in cosine distance to their students than a random same-field author does. Two open encoders reproduce the effect at $12.6\%$ and $14.5\%$. \emph{Academic siblings}, two students of one advisor who may never have met, sit $30.4\%$ closer across $8{,}360$ pairs, even when they studied at different institutions. Pairs who share only an institution and a field show negligible similarity. Given a closed-set attribution task over the same corpus, the system's errors occur on the true author's advisors and academic siblings $11$ times more often than chance.
Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
arXiv:2608.14843 [cs.CL]
(or arXiv:2608.14843v1 [cs.CL] for this version)
https://doi.org/10.48550/arXiv.2608.14843 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 261 words
