Dominik Straub

This a pretty funny encapsulation of ai scientist work. The authors hand it some softball software tasks, then it comes back with very different results than what it’s replicating and the authors shrug at the data being different

www.nature.com/articles/s43...

🚀🎓We have exciting opportunities at the PhD and Postdoc level, including in the new Simons Collaboration on Ecological Neuroscience, a 10-year initiative spanning a powerful international network of labs working across species and disciplines - please reach out if interested
bsky.app/profile/simo...

was at an event on AI for science yesterday, a panel discussion here at NeurIPS. The panelists discussed how they plan to replace humans at all levels in the scientific process. So I stood up and protested that what they are doing is evil.

Full post:
togelius.blogspot.com/2025/12/plea...

3. I appealed to the joy of doing science—and got the predictable question:

Would you rather live in a world where AI cured cancer or live in a world where you personally get to do science?

But that's not the dilemma we face at all. There's no cause to think that AI could do anything of the sort.

8000 Artikel über zu viel "Wokeness" an den deutschen Unis; vornehmes Schweigen dazu, dass diese Unis gerade im Rahmen einer harten Austeritätspolitik abgewickelt werden.

Tobias Niehues @tobnie.bsky.social is presenting our work with @dominikstrb.bsky.social 'Amortized Bayesian decision-making for inferring decision-making parameters from behavior' at the Amortized ProbML Workshop and the @ellis.eu UnConference. Please come by our poster!

I should have known better than to open this link at work :D

I used the first half of the course at half speed in a stats for psychology course and found that it worked really well.

If you're looking for a PhD position, check out this opportunity!

I had a great time doing my PhD in Darmstadt. Exciting research with lots of opportunities for collaboration across AI and cognitive science in a very livable city.

When will decisions be released to the reviewers? As a reviewer, I would be curious to see what happened to my batch of papers.

Q: Why did the LLM cross the road?

A: We're not sure, but it achieved 94.7% on CHIKENBench-Large

The vision of human-out-of-the-loop science fundamentally mistakes what knowledge production is. It aims at the production of, wait for it, knowledge: a thing that definitionally involves an epistemic agent or agents, unless you are theistic about it.

Cognition rears its ugly head during a meeting of psychophysicists

It's a bit of a prisoner's dilemma situation.

Academics are under unrelenting pressure to accept the narrative of the inevitability of #generativeAI & to embrace it in teaching & learning. We resist - because it is ecologically destructive, ethically corrupt, & because it undermines the thinking abilities that make us both human & intelligent.

New at the university: Prof. Weber! Osnabrück University welcomes Prof. Dr. Lilian Weber! Since August 15, she has held the professorship "Cognitive Modeling" at the School of Human Sciences. A warm welcome!

In other news, I am now a ✨ COXI ✨ (cognitive scientist in Osnabrück, Germany)

www.uni-osnabrueck.de/en/campus-li...

I'll be looking for PhD students & postdocs later this year, so watch this space if combining cognitive modelling, EEG & tFUS to study flexible cognition sounds exciting to you 🚀

G'day #ECVP!
Today, three more colleagues present their posters.

38: @celinehonekamp.bsky.social will introduce you to dual adaptation in the temporal domain.

48: Tarek will present a relative depth experiment conducted on the multifocal display he built.

An image showing the presentation times and places of the members of the Perception Lab at TU Darmstadt.

I won't get to join you all at #ecvp2025 @ecvp.bsky.social this year because I'm on holiday in Australia... but my lab is there in force! Please do go say hi, and check out my students' and postdocs' amazing work!

@ben.graphics @lschmittwilken.bsky.social @lreining.bsky.social

#ECVP2025 be ready: my poster on repeated exposure to delay in two different tasks is up on Thursday 15:30-17:00 (No 38)

Come and say hi

I'd be glad to see a lot of people at our poster on Tuesday, August 26th, 10-11.30am at #ECVP2025 and tell you more about which cost functions shape your behavior in sensorimotor decision-making!

Looking forward to meeting you #ECVP2025 Mainz this week, including collaborative work with @tobnie.bsky.social @dominikstrb.bsky.social @ookenfooken.bsky.social @fatatai.bsky.social @tsawallis.bsky.social @mamassian.bsky.social @guidomaiello.bsky.social @mariaeckstein.bsky.social and many others

my friend/colleague Frank Jäkel wrote a book on AI. I sadly don't know German but I happily know Frank, and I've heard him talking about this for a while now, and just on that basis I'd recommend the German speakers in the audience check it out

post 1 disappointed Weizenbaum for every thinkpiece u read like this

Upper left sketch shows the problem description the paper tackles, which is the decision-making problem that the subject needs to solve versus the inverse problem about the behavioral parameters that the researcher wants to infer. Bottom left sketch shows probabilistic graphical model of the behavior as formalized in our framework. Right panel shows the results of the paper. From top to bottom it shows example data, results of the model comparison, inferred cost functions and inferred prior beliefs of the subject. Five tasks are organized in columns by which cost function described subjects' behavior in the respective task best. We found three different cost functions, None of which are quadratic.

I'm presenting our work "Revisiting Cost Functions in Sensorimotor Decision-Making" at #CCN2025!

Stop by our poster (@dominikstrb.bsky.social‬, @c-rothkopf.bsky.social‬) and learn more about how to rethink common modeling assumptions.

📅 When: Friday, August 15, 2pm–5pm
📍 Where: De Brug, Poster C1

I have never understood the insane hostility that LLM fanboys have towards academics who disagree with them (see also: Gary Marcus). (Also how does “we fed an AI every available unit of human language and it’s pretty good” refute “10k words a day isn’t enough for babies to learn as well as they do”)

Delay adaptation does not transfer between discrete button press actions and continuous control - Loes CJ van Dam, Svenja Kernig, Karina Lazarova, Melisa Ünal, Nicole Gappa, Benjamin Straube, Thomas S... When interacting with technology, humans often deal with delays between an action and the desired action outcome. Through delay adaptation these delays will bec...

⬅️➡️ "Delay adaptation does not transfer between discrete button press actions and continuous control" by van Dam et al

Delay adaptation occurs for stylus (graphics tablet) and arrow buttons (keyboard) but doesn't transfer between response styles

journals.sagepub.com/doi/full/10....

Wahl der Verfassungsrichter: Brosius-Gersdorf berichtet von Drohungen nach gescheiterter Wahl Aufgrund von Drohungen hat Frauke Brosius-Gersdorf ihre Mitarbeiter gebeten, nicht am Lehrstuhl zu arbeiten. Einen Verzicht auf ihre Kandidatur schließt sie nicht aus.

After her party halted the election of Brosius-Gersdorf as a constitutional judge using fake news, the federal minister of science (and space!) expects candidates to have "a little resilience" and "being able to take criticism". Is this how to protect academic freedom?

www.zeit.de/politik/deut...

Wahl der Verfassungsrichter: Brosius-Gersdorf berichtet von Drohungen nach gescheiterter Wahl Aufgrund von Drohungen hat Frauke Brosius-Gersdorf ihre Mitarbeiter gebeten, nicht am Lehrstuhl zu arbeiten. Einen Verzicht auf ihre Kandidatur schließt sie nicht aus.

Eigentlich sollte eine Wissenschaftsministerin Stimme der Wissenschaft im politischen System sein.
Bär will hingegen was anderes sein und fordert von Wissenschaftlerinnen "Kritikfähigkeit" ein, wenn sie diffamiert werden.

www.zeit.de/politik/deut...

So much alignment talk ignores the more traditional and pressing question about who should own the means of production in the first place since that ultimately determines who gets to make these decisions

@meredithmeredith.bsky.social is such a fantastic public face of the countermovement to tech company overreach and creeping fascism.