Tessera
Folgen

Sebastian Sauppe

@sebastiansauppe.bsky.social

SNSF Ambizione fellow | «Cognitive & Developmental Neuroscience of Language» @ Developmental Psychology: Infancy and Childhood, Department of Psychology, University of Zurich | https://sites.google.com/site/sauppes

73 Posts517 Follower307 Folgt
Ben Williamson@benpatrickwill.bsky.social · 3w ago

The anatomy of an AI-supported academic article is becoming more apparent. I desk reviewed ~70 manuscripts this week for our journal. Reckon 20ish had significant AI assistance. On top of hundreds of others recently I'm starting to see patterns. Are other editors are seeing the same or similar? 1/

Deutsche Forschungsgemeinschaft@dfg.de · last mo.

Wie können wir die Wissenschaftsfreiheit gegen zunehmende Angriffe schützen? Ein neues Positionspapier einer DFG-Senats-AG zeigt auf, wie Resilienz von Wissenschaft & Wissenschaftssystem gestärkt werden kann & wie auch Politik, Wirtschaft & Gesellschaft unterstützen können: sohub.io/9tje

Grafik mit buntem abstraktem Papierwirbel auf dunkelblauem Hintergrund. Überschrift: Orientierungen und Empfehlungen - Positionspapier zur Stärkung der Freiheit und Resilienz der Wissenschaften. Foto: Adobe Stock
Sanktionsfrei@sanktionsfrei.bsky.social · last mo.

3,85 € pro Tag für Frühstück, Pausenbrot, Mittagessen und Abendbrot. Die Bundesregierung erwartet, dass Kinder damit satt werden und erfolgreich lernen können. 🔎Wer damit ein Schulkind durch den Tag bekommt, kann sich gerne in den Kommentaren melden...

Wer hungrig in die Schule geht, kann keine Leistung bringen. Schulkindern in der Grundsicherung stehen im Regelsatz 3,85 € pro Tag für alle Mahlzeiten zur Verfügung.
Holger Hestermeyer@hhesterm.bsky.social · last mo.

On my timeline I have seen suggestions that FIFA might have acted in an ethically dubious way. I am shocked, SHOCKED at the insinuation the organisation that awards the highly esteemed FIFA peace prize might not always maintain the highest ethical standards.

Arne Semsrott@arnesemsrott.bsky.social · last mo.

Bitte unterzeichnet unsere Petition gegen die IFG-Abschaffung! Das hilft uns dabei, Druck gegenüber der SPD aufzubauen. In den ersten 14 Stunden schon 120.000 Unterschriften!

FragDenStaat@fragdenstaat.de · last mo.

Die Spitzen von CDU, CSU und SPD wollen das Informationsfreiheitsgesetz abschaffen – einen Grundpfeiler unserer Demokratie. 2025 haben wir den Angriff der Union aufs IFG gemeinsam abgewehrt. Das schaffen wir auch heute – unterzeichnet jetzt unsere WeAct-Petition! weact.campact.de/petitions/sp...

Hannes Tschürtz@hannestschuertz.com · 2mo ago

Eigentlich müsste die ganze diplomatische Welt sehr sehr laut schreien. Untragbare Zustände schon bevor die WM beginnt. Ein echter Skandal. sport.orf.at/stories/3156...

FIFA WM 2026: Somalische Kritik nach verweigerter Einreise

Der für die Fußball-WM vorgesehene somalische Schiedsrichter Omar Artan ist nach seiner verweigerten Einreise in die USA „sehr enttäuscht“. Er sei „nur ein Referee“, der versuche, sich mit der WM-Teil...

sport.orf.at

Peter Cook@pfcook.bsky.social · 3mo ago

Great work from Simone Videsen. We found sperm whale slow clicks maintain extreme isochrony with inter-click intervals up to ~10s. That's beyond any known neural system for precise interval timing. May be a long-distance communication system. nyaspubs.onlinelibrary.wiley.com/doi/10.1111/...

NYAS Publications

In this study, we show that male sperm whales produce low-frequency slow clicks at low repetition rates with extreme source levels > 200 dB re 1 µPa (pp), making them the loudest mammalian communicat...

nyaspubs.onlinelibrary.wiley.com

Dan Kleinman@dankleinman.bsky.social · 3mo ago

“The replication crisis, if there is one, cannot be established by the methods used to declare it.” Really interesting, thought-provoking stuff. (And the callback in the paper title is perfect).

Berna Devezer@devezer.bsky.social · 3mo ago

📣 The difference between replicable and not replicable is not itself scientifically replicable. 📣 New work with Erkan Buzbas, showing that verdicts such as "X% of results replicated" are based on an inferential machinery that doesn't work. arxiv.org/abs/2604.26268

Screenshot of a paper's title page. Title: "The Difference Between 'Replicable' and 'Not replicable' is not Itself Scientifically Replicable". Authors: Berna Devezer and Erkan O. Buzbas, both at the University of Idaho — Devezer in the Department of Business and the Institute for Modeling Collaboration and Innovation, Buzbas in the Department of Mathematics and Statistical Science. Authors contributed equally. Corresponding author: bdevezer@uidaho.edu. Abstract: Replication studies estimate the replicability rate of scientific results by aggregating binary verdicts of experiments. Exact replications are rarely attainable, so most replication sequences are non-exact. Experiments differ in ways that matter and do not share a single common data-generating process. We formalize two statistical interpretations of this non-exactness. In a shared latent rate model (benchmark), experiments are exchangeable and depend on a common random replicability rate. In a conditionally independent rates model (operational), each experiment has its own replicability rate drawn independently from a population distribution. Under the shared latent rate model, even small variability among replicability rates induces an irreducible variance floor on the estimated mean replicability rate that cannot be eliminated by adding more replications. Under the conditionally independent rates model, the degree of non-exactness is not identifiable from standard replication data, because one binary verdict per experiment contains no information about between-experiment heterogeneity. Researchers therefore cannot tell which precision regime they are operating in or whether high- and low-replicability sequences can be distinguished in principle. As a result, the usual data structure of one binary verdict per experiment cannot support reliable demarcation between "replicable" and "not replicable" results and systematically understates uncertainty, making high- and low-replicability sequences appear discrim…
Mark Dingemanse@markdingemanse.net · 4mo ago

do people generally find this signature llm style of oversimplistic dichotomies compelling? still? because I see sth like Figure 1, with the cute monikers, the four quadrants, and the glib summaries, and my GenAI dosimeter ☢️ goes wild I am not surprised to see this line in the acknowledgements

A four-quadrant diagram discussing arguments and research programs related to language models. Top left: "The String Statistics Strawman" with a faulty argument that LMs cannot learn language and refutations. Top right: "The "As Good As It Gets" Assumption" with a faulty argument that LMs are useless for serious language science and refutations. Bottom left: "The Statistical Structure Steelman Research Program" with key research questions on processing, learning, structure, and real patterns. Bottom right: "The "Yes, And" Research Program" listing shortcomings of current LMs and corresponding research opportunities.Generative AI (Claude and ChatGPT) was used for brainstorming, summarizing, and finding relevant literature