Refraction Notch

Academic Sleuth

[Michael 英泉 Eisen @mbeisen]

Umm... @Nature's Impact Factor is in the Epstein Files.

[Refraction Notch @RefractionNotch]

Artificial intelligence researchers hit by flood of ‘slop’ https://t.co/LtI323AS1W

[Refraction Notch @RefractionNotch]

@mumumouse2 Fig. 10e seems to have two tiny backtrack instances.

[Refraction Notch @RefractionNotch]

@mumumouse2 It's not enough just to look around the corner. Without diving into the middle of the wild waves, you can hardly get to the anatomy of the problem.

[Refraction Notch @RefractionNotch]

@ImageTwinAI @ElsevierConnect @Proofig Hi,
If you haven't seen the paper yet, here is its DOI:

DOI: 10.1016/j.chemosphere.2025.144628

[Refraction Notch @RefractionNotch]

🧵 [1/2]
Dear Elsevier,
@ElsevierConnect
See this "Chemosphere" Retraction Notice
https://t.co/Qts42A8e44…
"After thorough analysis, these original SEM images did not reveal evidence of manipulation, nor signs of artificial generation through AI tools or similar means. However…

[Refraction Notch @RefractionNotch]

Dear Dr. Jekyll,

I am pleased to inform you that your paper has been accepted for publication.

Truly yours,
Dr. Hyde

[Refraction Notch @RefractionNotch]

@mumumouse2 That's an alien. 👽

[Sholto David @addictedtoigno1]

Reverse Image Forensics Challenge: Try to find a unique area in Figure 9 of this recent Scientific Reports paper: 10.1038/s41598-025-17456-6 [Aug 2025] - Annotated by ImageTwin

[Sholto David @addictedtoigno1]

I have written a blog at ForBetterScience. A German lab has published dozens of papers with images that have been scribbled on, or manipulated with the clone tool. Blogs can take a while to read, but maybe scroll through the pictures! Plenty to see! (link in the next comment)

[Refraction Notch @RefractionNotch]

@ElsevierConnect @ImageTwinAI @Proofig Oops! Sorry for the inconvenience. Here is the correct link:
https://t.co/X9mXp0lIvl
DOI: 10.1016/j.chemosphere.2025.144628

[Refraction Notch @RefractionNotch]

[2/2]
However,… It was noted that the specific sections identified as compromised in the published images correspond to blurry regions in the original data."

What "AI tools or similar means" do you use @ElsevierConnect?!

@ImageTwinAI @Proofig
#imageforensics

[WU Guangheng @Dr_5GH]

Abnormal data patterns are observed on the standard deviation values on the Table 3 of this article. Several numbers repeat too many times than expect, such as "0.89" repeats 30 times, "0.44" repeats 23 times. Such abnormal pattern suggests that the data may be manipulated.

[Refraction Notch @RefractionNotch]

@mumumouse2 They are doppelgänger.

[John B. Holbein @JohnHolbein1]

Look at the distribution of z-values from medical research!

[Sholto David @addictedtoigno1]

Something fishy about this figure🥁Scientific Reports September 2025. https://t.co/egrYED2H4x

[Refraction Notch @RefractionNotch]

A tide of fabricated research risks scientific integrity
https://t.co/1BXeubPAmM

[Retraction Watch @RetractionWatch]

Researcher “repulsed by the prevalence of scientific misconduct” says “science itself could be a form of dissent.”
https://t.co/Kd8erHfQ67

[WU Guangheng @Dr_5GH]

Lego-like bar charts, and other issues on the article. Please find the details: https://t.co/tzk6hH0Y08

[Refraction Notch @RefractionNotch]

Systematic fraud uncovered in mathematics publications

Source: https://t.co/S3fRuliDzz
https://t.co/uUncWx9soo