Category: arXiv

An expert in AI at Google has admitted he used the technology to help write a preprint manuscript that commenters on PubPeer found to contain a slew of AI-generated phrases like “squared blunder” and “info picture.” 

The paper, “Leveraging GANs For Active Appearance Models Optimized Model Fitting,” appeared on arXiv.org in January but was withdrawn April 7. The author, Anurag Awasthi, is an engineering lead in AI infrastructure at Google. In a PubPeer comment, he described the paper as a “personal learning exercise.” 

In March 2025, sleuth Guillaume Cabanac, creator of the Problematic Paper Screener, pointed out in a PubPeer comment the paper included several tortured phrases. These phrases indicate AI use and occur when large language models try to find synonyms for common phrases. In Awasthi’s paper, “linear regression” became “straight relapse,” and “error rate” became “blunder rate,” among others. 

Continue reading ‘Squared blunder’: Google engineer withdraws preprint after getting called out for using AI

While using bibliometric techniques to measure how disruptive research papers are to their field of study, Robin Haunschild and Lutz Bornmann stumbled across a strange phenomenon. 

Just under 45,000 academic papers contained citations to themselves, they found. Haunschild and Bornmann — both information scientists at the Max Planck Institute for Solid State Research in Stuttgart, Germany — found these “paper self-citations” in journals indexed by Clarivate’s Web of Science since 1980. 

Some 7,943 different journals had at least one self-citing paper, the researchers report in their study, posted on arXiv.org earlier this month. Eight journals alone covered 10% of the sample papers, and 129 publications covered the top third. More than 31,000 of the papers appeared under the ‘article’ category in Web of Science, followed by just over 6,000 listed as ‘corrections’ and just under 2,500 as ‘reviews.’

Continue reading Why do nearly 45,000 scholarly papers cite themselves?

When we last heard from Eve Armstrong one year ago today, she was a postdoc at the BioCircuits Institute at the University of California, San Diego, musing mathematically about what would have happened if she had asked one Barry Cottonfield to her high school prom in 1997. Today, she is a postdoc at the Computational Neuroscience Initiative at the University of Pennsylvania — where, it would appear from a new preprint, titled “Colonel Mustard in the Aviary with the Candlestick: a limit cycle attractor transitions to a stable focus via supercritical Andronov-Hopf bifurcation,” she has solved a brutal murder.

To be more precise:

We establish the means by which Mr. Boddy came to transition from a stable trajectory within the global phase space of Philadelphia, Pennsylvania to a stable point on the cement floor of an aviary near the west bank of the Schuylkill River.

The preprint will be posted at arXiv tomorrow, but Armstrong did not want to keep the world in suspense any longer, so we are hosting the manuscript here at Retraction Watch on a more appropriate day, April 1. She was also kind enough to answer a few questions about the work: Continue reading Clue fans, here’s scientific proof that it was Colonel Mustard with a candlestick