Post

Andrew White 🐦‍⬛
@andrew.diffuse.one

Image duplication has been a powerful signal for detecting scientific fraud, but is irrelevant in many fields. I've been working a bit on finding new signals like it that work across fields. I've found one using LLMs that can predict retractions, weakly, for $1 per paper. 1/4

Andrew White 🐦‍⬛
@andrew.diffuse.one

What is a universal way to check for signs of fraud in a paper? I investigated faithfulness of citations – are citations consistent with cited sources, are they irrelevant? This does significantly correlate with if a paper is subsequently retracted 3/4

Andrew White 🐦‍⬛
@andrew.diffuse.one

This is still an early topic and my work is very preliminary, but I think we may be able to start auditing scientific literature at scale. I’ve written up a lot of thoughts, background, and analysis in a blog post: diffuse.one/p/d1-008 4/4

0 reposts1 like

Stijn De Baerdemacker
@cortogantese.bsky.social

This is an exciting project. I see a potential use in developing a tool for "fair citation", in which digging into a network grown from a few seed citations might unveil a hidden discussion about one's work in the very early stages of said work. Again, .pdf retrieval will be the bottleneck