Sprogmodeller skal blive bedre til sprog 🗣️
Tillykke til ITU-lektor Rob van der Goot, der med støtte fra @carlsbergfondet.dk vil undersøge, hvordan man kan integrere sproglig viden direkte i sprogmodellers design.
Det udfordrer den dominerende tilgang, der primært bygger på data og beregningskraft.
Lucky University of Toronto! Congratulations to them and to you!
19.12.2025 10:18 — 👍 2 🔁 0 💬 0 📌 0
Slide reading: Our alrogithms are not yet user-friendly. Innoexplorer solves this: Backend developer streamlines the algorithms, creates usable software; web developers visualize our solutions on a web platform.
Michael, Anastassia and Manuel win Innoexplorer Grant!
nerds.itu.dk/2025/12/11/m...
Today, social media platforms hold the sole power to study the effects of feed-ranking algorithms. We developed a platform-independent method that reranks participants’ feeds in real time and used this method to conduct a preregistered 10-day field experiment with 1256 participants on X during the 2024 US presidential campaign. Our experiment used a large language model to rerank posts that expressed antidemocratic attitudes and partisan animosity (AAPA). Decreasing or increasing AAPA exposure shifted out-party partisan animosity by more than 2 points on a 100-point feeling thermometer, with no detectable differences across party lines, providing causal evidence that exposure to AAPA content alters affective polarization. This work establishes a method to study feed algorithms without requiring platform cooperation, enabling independent evaluation of ranking interventions in naturalistic settings.
New paper in Science:
In a platform-independent field experiment, we show that reranking content expressing antidemocratic attitudes and partisan animosity in social media feeds alters affective polarization.
🧵
I am hiring a postdoc for a DFF-funded project on social influence, and the decision processes that fuel rich-get-richer dynamics in the online/offline world. The position is for up to a year, competitive Danish salary, remote work possible. Interested or know somebody? DM me or share!
01.12.2025 08:29 — 👍 18 🔁 24 💬 1 📌 0
Portrait photo of a smiling researcher, grey background
Vedran wins Independent Research Fund Denmark Grant!
nerds.itu.dk/2025/11/03/v...
It was a ton of fun to play a small role in this project lead by the brilliant Marta Ewa Lech & @jonassjuul.bsky.social. A lot of drama & change in the world of music reflected in the song-dynamics of the Billboard Hot 100!
22.09.2025 09:27 — 👍 5 🔁 3 💬 0 📌 0
Extra: Also, thanks to Alice Nadeau. Alice brought to my attention that some artists have recently been arguing that it is getting harder to make a hit (see e.g., www.bbc.com/news/enterta...). Our conversations sparked the idea for this project.
22.09.2025 09:07 — 👍 1 🔁 0 💬 0 📌 0
7/7
This was a fun project to work on and write up. Thanks to the great Marta Ewa Lech and the always-wonderful @sunelehmann.com .
6/7
Our results are especially interesting in the context of collective attention: Whereas books, news stories, movies, etc. get popular, then lose popularity at a higher and higher pace (see www.nature.com/articles/s41...), the most-popular hit songs now stay popular for longer.
4/7
On today’s chart, there are
- Fewer songs by new artists
- More songs performed by established ‘hitmakers’ (but these songs reach lower positions on average than used to be the case for hitmakers)
- Many, many more collaborations than previously.
3/7
- Song chart lifetimes peaked in the 2000s, then dropped.
- In the past two decades, lifetime inequality has surged (gini >0.6), because
- Mega hits are now staying on the chart for months, even years;
- Churn in the bottom is huge.
(tales of both the winner-takes-all and long tail theories)
2/7
The Billboard Hot 100 has tracked US music consumption since 1958.
In her master’s thesis, the awesome Marta Ewa Lech studied all chart entries to understand how song and artist success has evolved since the chart’s creation.
New paper out in EPJ Data Science!
We analyze 65 years of Billboard Hot 100 data to ask:
🎵Are songs moving differently up and down the chart nowadays?
📉 Are hits shorter-lived?
📈 Or do superstars dominate more than ever?
Paper: epjdatascience.springeropen.com/articles/10....
🧵 1/7
I am ecstatic to join the Danish Young Academy starting September 2025.
I cannot wait to meet the other members, start new collaborations, and learn more about the new opportunities and challenges that lie at the intersection of research, universities, and society in Denmark.
New paper alert. "More is different" & the LLMs.
LLMs are usually studied in isolation. But what happens when they start interacting? We explored this by looking at their collective behaviour.
Work with @ariel-flint.bsky.social and @lajello.bsky.social
1/
The new pope has a bachelor degree in mathematics! And I thought the job prospects could not get any better for math majors!
08.05.2025 19:33 — 👍 0 🔁 0 💬 0 📌 0
Photograph of a smiling researcher
Luca Maria Aiello becomes Full Professor
nerds.itu.dk/2025/04/01/luca-m
Congratulations Luca @lajello.bsky.social, well deserved! 🎉👨🏫
We have several open positions, both Phd & Postdoc: nerds.itu.dk/positions/
On the topics of epidemic preparedness, and network science & AI & #ScienceOfScience #SciSci
Deadlines in April and May!
📢 Call for PhD and Postdocs: Using data science to improve epidemic preparedness
nerds.itu.dk/2025/03/25/c...
We are looking for Phds and Postdocs to join @jonassjuul.bsky.social at @itu.dk in Copenhagen improving inference, forecasting, and mitigation in future pandemics. Deadline May 1st.
3 relevant papers:
- www.nature.com/articles/s41...
- www.nature.com/articles/s41...
- journals.aps.org/pre/abstract...
I am hiring PhD students and Postdocs to join me in beautiful Copenhagen. Together, we will develop data science methods to improve epidemic preparedness.
Copenhagen is amazing, salary is good, we have plenty of funding, and a great community.
Read more..: candidate.hr-manager.net/ApplicationI...
There are days in life that shake you.
I’m shattered 💔 to share that I just found out that the US Government terminated my 2024 NIH Director’s Early Independence Award (~$2 million), threatening my long-promised assistant professor job at Columbia University
& academic career... 1/🧵
Map showing a highway section in red and social ties in space crossing the highway. Wherever a tie crosses the highway, there is a cross. There are 94 crosses.
🎉 New paper in PNAS: Urban highways are barriers to social ties
https://www.pnas.org/doi/10.1073/pnas.2408937122
Highways are barriers that cut opportunities for social ties. We quantify this effect by overlaying the US highway network with millions of social ties from Twitter.
New NERDS paper on academic mentorship
nerds.itu.dk/2025/03/03/n...
Jonas wins Novo Nordisk Data Science Investigator Grant! 🥳
nerds.itu.dk/2025/02/26/j...
