@resurgequality.bsky.social
@bengoldacre.bsky.social Hey I once left this Naive comment on your turf about using LLM's to add new functions to Opensafely. I read about prompt injections rn and realised it could work pretty well if you button it up: Limit it to single useful functions that are pressed by a button= No prompts.
December 10, 2025 at 11:35 PM
@bengoldacre.bsky.social Hey I once left this Naive comment on your turf about using LLM's to add new functions to Opensafely. I read about prompt injections rn and realised it could work pretty well if you button it up: Limit it to single useful functions that are pressed by a button= No prompts.
Reposted
Come on man
The FDA plans to begin requiring one clinical study, instead of the standard two, for medical products before consideration for approval. www.statnews.com/2025/12/04/f...
FDA to lower number of trials required for approval of drugs, other medical products
The FDA plans to begin requiring one clinical study, instead of the standard two, for medical products before consideration for approval.
www.statnews.com
December 5, 2025 at 1:49 AM
Come on man
@dereklowe.bsky.social Now i got an RSS reader only because I could have easily missed your Reverse Proteolysis Blog post from only rrading Bluesky. This thing was insane news. It made me think of prions as well..
December 4, 2025 at 11:55 PM
@dereklowe.bsky.social Now i got an RSS reader only because I could have easily missed your Reverse Proteolysis Blog post from only rrading Bluesky. This thing was insane news. It made me think of prions as well..
This "looks like" you could just throw it into the AlphaGo @googledeepmind.zpravobot.news.ap.brid.gy engine and change the rules in how the boars looks a bit to get a potential improvement.
The traveling salesperson problem is one of the most notorious, longstanding problems in theoretical computer science. 🧵
December 4, 2025 at 10:33 PM
This "looks like" you could just throw it into the AlphaGo @googledeepmind.zpravobot.news.ap.brid.gy engine and change the rules in how the boars looks a bit to get a potential improvement.
Reposted
Random colleagues in other disciplines now commonly say "the code's on OSF". It's just so great to see an academic org like COS, that goes beyond advocating for change, and actually builds the tools and services that make science work better. More!
Each year, hundreds of people donate to COS, helping keep OSF free and strengthening our open scholarship training and community support across research fields. This #GivingTuesday, we’d be grateful for your support: www.cos.io/support-yef25.
December 2, 2025 at 8:28 PM
Random colleagues in other disciplines now commonly say "the code's on OSF". It's just so great to see an academic org like COS, that goes beyond advocating for change, and actually builds the tools and services that make science work better. More!
Reposted
Scientific error-checking, difficulty: easy
Find some issues in this table of results from an RCT of vitamin D for COVID-19. There's a couple of obvious issues, and some that are less easy to spot.
Find some issues in this table of results from an RCT of vitamin D for COVID-19. There's a couple of obvious issues, and some that are less easy to spot.
November 25, 2025 at 4:59 AM
Scientific error-checking, difficulty: easy
Find some issues in this table of results from an RCT of vitamin D for COVID-19. There's a couple of obvious issues, and some that are less easy to spot.
Find some issues in this table of results from an RCT of vitamin D for COVID-19. There's a couple of obvious issues, and some that are less easy to spot.
Reposted
Hey @nature.com, have you got an explanation for how the hell THIS happened? & especially why you accepted a paper with such a bizarre piece of genAI slop in it?!
& more to the point, why we should take you seriously at all going forward?
www.nature.com/articles/s41...
& more to the point, why we should take you seriously at all going forward?
www.nature.com/articles/s41...
November 27, 2025 at 3:03 PM
Hey @nature.com, have you got an explanation for how the hell THIS happened? & especially why you accepted a paper with such a bizarre piece of genAI slop in it?!
& more to the point, why we should take you seriously at all going forward?
www.nature.com/articles/s41...
& more to the point, why we should take you seriously at all going forward?
www.nature.com/articles/s41...
@lessig.bsky.social youtu.be/-NuXpwB2DV4?.... Johnny Harris finally did a pretty good piece on this. You probably already saw it.
How billionaires stole America's elections
YouTube video by Johnny Harris
youtu.be
November 26, 2025 at 2:24 AM
@lessig.bsky.social youtu.be/-NuXpwB2DV4?.... Johnny Harris finally did a pretty good piece on this. You probably already saw it.
Reposted
lads i found ANOTHER Table of Impossible Summary Statistics in a "green economics" paper 🚩
Can you see the problem?
Can you see the problem?
November 22, 2025 at 2:29 PM
lads i found ANOTHER Table of Impossible Summary Statistics in a "green economics" paper 🚩
Can you see the problem?
Can you see the problem?
Reposted
November 18, 2025 at 6:57 PM
Reposted
In dieser Studie glaubten 58% aller Jugendlichen nach einer Mental Health Awareness Intervention #ADHS zu haben. "Nur" 28% glaubten dies vorher. Keine*r von ihnen erfüllte die offiziellen Kriterien.
www.madinamerica.com/2025/11/adhd...
www.madinamerica.com/2025/11/adhd...
ADHD Awareness Campaigns Lead to Huge Jump in False Self-Diagnoses
More than half of young adults began to think they had ADHD after awareness campaign—despite not meeting criteria for this disorder.
www.madinamerica.com
November 19, 2025 at 8:24 AM
In dieser Studie glaubten 58% aller Jugendlichen nach einer Mental Health Awareness Intervention #ADHS zu haben. "Nur" 28% glaubten dies vorher. Keine*r von ihnen erfüllte die offiziellen Kriterien.
www.madinamerica.com/2025/11/adhd...
www.madinamerica.com/2025/11/adhd...
Reposted
EV: Rejection rate now 57% (48% desk reject by RI team). We use an AI model that can recognize paper mill papers based on 15 data points (red flags/trust markers), trained on a 64K dataset.
Benchmarking AIRA with other tools - but not much overlap. Still learning from this.
#IRICSydney
Benchmarking AIRA with other tools - but not much overlap. Still learning from this.
#IRICSydney
November 18, 2025 at 3:27 AM
EV: Rejection rate now 57% (48% desk reject by RI team). We use an AI model that can recognize paper mill papers based on 15 data points (red flags/trust markers), trained on a 64K dataset.
Benchmarking AIRA with other tools - but not much overlap. Still learning from this.
#IRICSydney
Benchmarking AIRA with other tools - but not much overlap. Still learning from this.
#IRICSydney
Reposted
* Universities are worried about reputational risks. They will resist because a national ORI might mean they cannot cover up cases as much as before.
* Do not adapt the Dutch national ORI model! It just oversees institutional investigations, but cover-ups still happen.
#IRICSydney
* Do not adapt the Dutch national ORI model! It just oversees institutional investigations, but cover-ups still happen.
#IRICSydney
November 17, 2025 at 4:16 AM
* Universities are worried about reputational risks. They will resist because a national ORI might mean they cannot cover up cases as much as before.
* Do not adapt the Dutch national ORI model! It just oversees institutional investigations, but cover-ups still happen.
#IRICSydney
* Do not adapt the Dutch national ORI model! It just oversees institutional investigations, but cover-ups still happen.
#IRICSydney
Reposted
There is a lot of fuss today over whether chatbots can replace human participants in social sciences research when the solution is obvious: ask chatbots to simulate the views of social scientists and survey them on attitudes towards chatbots as substitutes for human subjects.
November 10, 2025 at 10:45 PM
There is a lot of fuss today over whether chatbots can replace human participants in social sciences research when the solution is obvious: ask chatbots to simulate the views of social scientists and survey them on attitudes towards chatbots as substitutes for human subjects.
The nonsense of bad/fake studies is the incentive to publish them: if I (am an applied material scientist &) want to keep my experience valuable, a good bet is to "help" others to publish fake but plausible papers for a small fee.
Experience from practice will stay relevant.
Experience from practice will stay relevant.
November 12, 2025 at 1:31 AM
The nonsense of bad/fake studies is the incentive to publish them: if I (am an applied material scientist &) want to keep my experience valuable, a good bet is to "help" others to publish fake but plausible papers for a small fee.
Experience from practice will stay relevant.
Experience from practice will stay relevant.
Reposted
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.ai
November 11, 2025 at 1:49 AM
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.ai
Reposted
Peter Doshi, who lead the RIAT initiative, writes about the recent legal actions on the 2001 publication of Study 329.
Never heard about Study 329 or RIAT?
Then I recommend reading this to understand the darker sides of EBM.
www.bmj.com/content/391/...
Never heard about Study 329 or RIAT?
Then I recommend reading this to understand the darker sides of EBM.
www.bmj.com/content/391/...
Notorious GSK paroxetine adolescent depression trial is slapped with expression of concern amid legal action
A long criticised study on antidepressant use in adolescents has been flagged with an “expression of concern” by a leading academic journal after the launch of legal action.12
George Murgatroyd III, ...
www.bmj.com
November 4, 2025 at 5:37 PM
Peter Doshi, who lead the RIAT initiative, writes about the recent legal actions on the 2001 publication of Study 329.
Never heard about Study 329 or RIAT?
Then I recommend reading this to understand the darker sides of EBM.
www.bmj.com/content/391/...
Never heard about Study 329 or RIAT?
Then I recommend reading this to understand the darker sides of EBM.
www.bmj.com/content/391/...
Reposted
November 1, 2025 at 8:19 PM
Reposted
arXiv will no longer accept review articles and position papers unless they have been accepted at a journal or a conference and complete successful peer review.
This is due to being overwhelmed by a hundreds of AI generated papers a month.
Yet another open submission process killed by LLMs.
This is due to being overwhelmed by a hundreds of AI generated papers a month.
Yet another open submission process killed by LLMs.
Attention Authors: Updated Practice for Review Articles and Position Papers in arXiv CS Category – arXiv blog
blog.arxiv.org
November 1, 2025 at 5:28 PM
arXiv will no longer accept review articles and position papers unless they have been accepted at a journal or a conference and complete successful peer review.
This is due to being overwhelmed by a hundreds of AI generated papers a month.
Yet another open submission process killed by LLMs.
This is due to being overwhelmed by a hundreds of AI generated papers a month.
Yet another open submission process killed by LLMs.
Reposted
@bmj.com Please look at PubPeer comments on an article you published last week. pubpeer.com/publications...
I think your research integrity dept shld act swiftly on this one, given clinical significance.
I'm aware of even more evidence of problems so let me know if this is not sufficient.
I think your research integrity dept shld act swiftly on this one, given clinical significance.
I'm aware of even more evidence of problems so let me know if this is not sufficient.
PubPeer - Prevention of acute myocardial infarction induced heart fail...
There are comments on PubPeer for publication: Prevention of acute myocardial infarction induced heart failure by intracoronary infusion of mesenchymal stem cells: phase 3 randomised clinical trial (P...
pubpeer.com
November 2, 2025 at 3:02 PM
@bmj.com Please look at PubPeer comments on an article you published last week. pubpeer.com/publications...
I think your research integrity dept shld act swiftly on this one, given clinical significance.
I'm aware of even more evidence of problems so let me know if this is not sufficient.
I think your research integrity dept shld act swiftly on this one, given clinical significance.
I'm aware of even more evidence of problems so let me know if this is not sufficient.
Reposted
In which we all learn the phrase “ballistic microscopy”:
Blasting Through Cells
www.science.org
October 29, 2025 at 4:48 PM
In which we all learn the phrase “ballistic microscopy”:
I said business therefore I also said fascism? People who can't read or think.
"unis are bad, therefore more fascism not less"
these fucking people
these fucking people
Read the researchers contracts and apinoff deals, or the way universities hoard real estate. Science is often a business theee days. Even rekigion is. And those who make no business with it even try to emulate the business aspect.
October 26, 2025 at 3:52 PM
I said business therefore I also said fascism? People who can't read or think.
Reposted
Very important article - I can't see how this is much better than MAHA fishing for dubious causes of autism except that the "microbiome" is currently more in fashion. And imagine what £50M could do for research on what interventions help autistic kids most.
New blogpost on my concerns about Wellcome LEAP new $50m program on autism/microbiome links. deevybee.blogspot.com/2025/10/a-le...
#autism #microbiome #biomarkers #diagnostic
#autism #microbiome #biomarkers #diagnostic
A LEAP into the future, or off a cliff: Wellcome LEAP's new $50M program
A few days ago, I saw this post on LinkedIn: How does the gut microbiome shape early brain development? That’s what FORM, a new $50 million...
deevybee.blogspot.com
October 22, 2025 at 6:38 AM
Very important article - I can't see how this is much better than MAHA fishing for dubious causes of autism except that the "microbiome" is currently more in fashion. And imagine what £50M could do for research on what interventions help autistic kids most.
Reposted
Wait, disordered proteins can be enzymes, too? It's a confusing world for sure.
Disordered Enzymes, Oh Dear
www.science.org
October 17, 2025 at 8:23 PM
Wait, disordered proteins can be enzymes, too? It's a confusing world for sure.