Tadeas Cely
56 posts

Tadeas Cely
@CelyTadeas
Postdoc in political science (Aarhus) Working on ideology, polarization, and how to save democracy from ourselves
Aarhus, DK Katılım Kasım 2018
338 Takip Edilen145 Takipçiler
Sabitlenmiş Tweet

Check my new article in @PolBehavior! doi.org/10.1007/s11109…
You can also read the full 🧵here:
muni.cz/go/pobe-article
English

In case you missed it, I have a new piece out in EJPR! :) doi.org/10.1111/1475-6…
EJPR journal@EJPRjournal
🆕 Are all #partisan camps equally ideologically aligned? 🌍 Using @ESS_Survey data from 131 parties in 15 European countries, @CelyTadeas finds there is a broad & substantive asymmetry ↔️ between the right & the left across ideological dimensions. 👉 bit.ly/4jrcCw5 #OA
English

@PRL_Tweets This is the result of an ongoing collaboration with my tremendous colleagues: Michal Toth, @SpacPeter , and
@MiroslavNemcok
English

@PRL_Tweets 🔗 See the full report of descriptive results here: osf.io/ju8fx
English

It was a thrill to be a visitor at @IPZ_ch and ETH Zurich. Thank you to everyone who made my stay so pleasant. I hope we will have the chance to meet again soon.

English

The Czech Fulbright Committee asked me to write a short blog piece about my stay—so here it is! Huge thanks to @PRL_Tweets for hosting me!
czechfulbright.blogspot.com/2024/11/tadeas…
English
Tadeas Cely retweetledi

New data: Half of Americans are *not confident* Harris will concede if she loses (51.6%), even fewer are confident Trump will concede (28.8%).
Republicans have more confidence in a Trump concession (58.8%) than a Harris concession (38.3%). Compared to 13.3% and 67.3% among Democrats.
Incredible how little confidence Americans have in our candidates

English
Tadeas Cely retweetledi

Just published in @Nature!
Research by @GoogleDeepMind including Nobel Laureate @demishassabis introduces a novel technique for reliably identifying LLM-generated content.
At its core, this approach subtly alters the next-token sampling process to introduce minor adjustments into the distribution of generated text.
These slight modifications embed a “signature” in the LLM-generated text that would be nearly invisible to a human reader but detectable using specific statistical tools.
A major advantage of this method is its efficiency: detection doesn’t require resource-intensive computations or direct access to the underlying LLM, which is often proprietary.
Full paper: nature.com/articles/s4158…

English

Only one week left to apply! Send your PAP or extended abstract our way and join us in Brno!
Tadeas Cely@CelyTadeas
🚨 Call for Papers 🚨 Come and join as at @MUNI_FSS in Brno to figure out consequences of political polarization. @seanjwestwood, @MarkusWagnerAT, @inpsy_muni, @LenkaHrbkov Help us reach more younger scholars by sharing this post.
English

Forecasting we've all been waiting for...🙃science.org/content/articl…
English
