‘Tortured phrases’, lost in translation: Sleuths find even more problems at journal that just flagged 400 papers

What do subterranean insect provinces and motion to clamor have to do with microprocessors and microsystems?

That’s an excellent question. Read on, dear reader.

Earlier this month, as we reported, Elsevier announced that it had concerns about some 400 papers published in special issues in one of its journals. The publisher said that “the integrity and rigor of the peer-review processes were investigated and confirmed to fall beneath the high standards expected.”

It seems the problems with the journal, Microprocessors & Microsystems, don’t stop there. A trio of researchers in France and Russia say they’ve found a host of other red flags in M & M, including signs of plagiarism, images lifted from other publications and evidence of “tortured phrases” suggestive of paper mills at work.

Tortured phrases are what happens to words that get translated from English into a foreign language, then back to English — perhaps by a computer trying to generate a scholarly publication for a group of unscrupulous authors. 

In a new preprint posted to the arXiv server, Guillaume Cabanac, a computer scientist at the University of Toulouse, and his colleagues — Cyril Labbé, of University Grenoble-Alpes and Alexander Magazinov, of Skoltech, in Moscow — describe their findings of tortured phrases in M & M and other journals. Labbe and Cabanac have been involved in a number of efforts that have uncovered likely products of SCIgen, which creates gibberish papers, and other problems in the literature.

Per the abstract: 

Probabilistic text generators have been used to produce fake scientific papers for more than a decade. Such nonsensical papers are easily detected by both human and machine. Now more complex AI-powered generation techniques produce texts indistinguishable from that of humans and the generation of scientific texts from a few keywords has been documented. Our study introduces the concept of tortured phrases: unexpected weird phrases in lieu of established ones, such as ‘counterfeit consciousness’ instead of ‘artificial intelligence.’ We combed the literature for tortured phrases and study one reputable journal where these concentrated en masse. Hypothesising the use of advanced language models we ran a detector on the abstracts of recent articles of this journal and on several control sets. The pairwise comparisons reveal a concentration of abstracts flagged as ‘synthetic’ in the journal. We also highlight irregularities in its operation, such as abrupt changes in editorial timelines. We substantiate our call for investigation by analysing several individual dubious articles, stressing questionable features: tortured writing style, citation of non-existent literature, and unacknowledged image reuse. Surprisingly, some websites offer to rewrite texts for free, generating gobbledegook full of tortured phrases. We believe some authors used rewritten texts to pad their manuscripts. …

Some of the mangled phrases the authors identified include:

• Profound neural organization for deep neural network 
• Motor vitality for kinetic energy
• Motion to clamor for signal to noise
• Discourse acknowledgement for voice recognition
• Ad our favorite: Subterranean insect province (or area, state, region or settlement — your choice) for ant colony

Cabanac told Retraction Watch: 

We noticed a huge number of questionable papers in the volumes and queue of the journal (we first encountered a sensible number – ca. 30 – papers with “tortured phrases”, then, upon a closer look, realized that the problem is even bigger).

He says that the journal has been silent in response to the concerns:

I’m afraid we got no reply to date, not even a ‘mail received’ acknowledgement.

We emailed Elsevier for comment but have yet to hear back.

Update, 1140 UTC, 7/21/21: An Elsevier spokesperson gave us this statement:

The Editor in Chief of Microprocessors & Microsystems has serious concerns about the integrity and rigor of the peer-review processes that led to the publication of over 400 papers in six Special Issues, as previously reported. Investigation and due process regarding individual articles published in these Special Issues is ongoing. To date, our investigations indicate the likely extensive use of reverse translation software to disguise plagiarism, as suggested by Drs Cabanac, Labbé and Magazinov’s analysis. We appreciate the authors helping bring this issue to light. We will provide further updates in accordance with Elsevier and the Committee on Publication Ethics best practices once the investigations have concluded.

Like Retraction Watch? You can make a one-time tax-deductible contribution or a monthly tax-deductible donation to support our work, follow us on Twitter, like us on Facebook, add us to your RSS reader, or subscribe to our daily digest. If you find a retraction that’s not in our database, you can let us know here. For comments or feedback, email us at [email protected].