A new machine learning system has flagged more than 250,000 cancer research papers that may be connected to so called “paper mills.”
The study, published in The BMJ, examined 2.6 million cancer research papers released between 1999 and 2024. It was led by QUT researcher Professor Adrian Barnett, from the School of Public Health and Social Work and Australian Centre for Health Services and Innovation (AusHSI), together with an international group of collaborators.
The researchers found that more than 250,000 papers contained writing patterns similar to those seen in studies that had already been retracted over suspected fabrication.
“Paper mills are companies that sell fake or low-quality scientific studies. They are producing ‘research’ on an industrial scale, and our findings suggest the problem in cancer research is far larger than most people realized,” Professor Barnett said.
How Paper Mills Produce Fake Research
Paper mills sell authorship positions and, in some cases, complete ready made scientific papers. These studies may contain reused text, unusual or awkward language, and fabricated data or images.
“Most likely, they’re relying on boilerplate templates which can be detected by large language models that analyze patterns in texts,” Professor Barnett said.
To search for these patterns, Barnett and his colleagues trained a language model called BERT. The system was designed to identify subtle textual “fingerprints” that repeatedly appear in known paper mill products.
When evaluated using verified examples, the model correctly detected suspicious papers 91 percent of the time.
“We’ve essentially built a scientific spam filter,” Professor Barnett said.
“Just like your email system can spot unwanted messages, our tool flags papers that match the writing style and structure we see in retracted, fraudulent work.”
Suspicious Cancer Papers Have Surged
The large scale analysis revealed several major trends:
Flagged papers rose sharply over the past two decades, increasing from about 1 percent in the early 2000s to a peak of more than 16 percent in 2022.
The suspected problem appears across thousands of journals published by major companies, including journals with strong reputations and high impact.
Suspicious papers were especially common in areas such as molecular cancer biology and early stage laboratory research.
Certain cancer types, including gastric, liver, bone and lung cancer, had particularly high rates of flagged studies.
Journals Begin Testing the AI Tool
Three scientific journals are already testing the system as part of their editorial review process. The goal is to help editors identify potentially fabricated manuscripts before they are sent to outside experts for peer review.
The researchers also plan to adapt the tool for use in other scientific fields. They expect its accuracy to improve as more confirmed examples of paper mill activity become available.
However, the team emphasized that papers identified by the system should not automatically be treated as fraudulent. The results are warning signals, not confirmed findings of misconduct, and each case still needs to be reviewed by human experts.
Why Fake Research Can Harm Patients
“Cancer research influences clinical trials, drug development and patient care,” Professor Barnett said.
“If fabricated studies make their way into the evidence base, they can mislead real scientists and ultimately slow progress for patients. That’s why it’s vital we get ahead of this problem.”