Oransky's right. 7 years to retract, and by now those papers are already in meta-analyses and reviews nobody's going back to fix. Journals move at review speed while mills keep shipping.

The Pollard case looks like a reference verification problem.

Problematic Paper Screener by Cabanac and Labbé scans 130M+ papers with 1000+ retractions triggered. I run a smaller open-source tool doing similar checks.

The question is why publishers don't integrate these at submission?

On deception: only ~5% of citations to retracted papers mention the retraction. Formally everything is done, but in the literature the paper keeps getting cited as normal. The gap is huge.

So universities with retracted top scientists are more likely to have integrity pages than clean ones? That's like reactive, not proactive compliance :(

Both approaches are important. Legal framing creates consequences, bug bounties give people a reason to actually look. Right now almost nobody gets paid for finding errors in published papers, so they just pile up.

Hi! I'm Jan, technical product manager. Recently built my first open-source tool - it helps researchers check their references for retracted papers. Just joined Bluesky.