We had such a fun time at UZH/ETHs Scientifica! People really seemed to enjoy our booth about Deepfakes 🥳 @evavogel.bsky.social @larakobilke.bsky.social

That's a wrap on #COMPTEXT2026! Huge thanks to lead organiser @chrisguarnold.bsky.social and IDAI/@unibirmingham.bsky.social team, plus all tutors, chairs, discussants & sponsors. Conference dinner at a world-renowned cricket stadium—what a treat. Safe travels home; see you at COMPTEXT 2027!

Looking for something to read after the long week? Check out our new preprint "Fact-Checks Can Help Inoculate LLMs Against Disinformation"! osf.io/preprints/so... Read a brief summary below -> /1

📊 New preprint with @evavogel.bsky.social @morganwack.bsky.social @esserfrank.bsky.social: we tested whether you can batch-classify multiple texts in a single LLM prompt without losing coding accuracy. Short answer: yes, up to about 100 items per prompt. Paper: arxiv.org/abs/2604.03684

Researchers waste 80% of LLM annotation costs by classifying one text at a time

Large language models (LLMs) are increasingly being used for text classification across the social sciences, yet researchers overwhelmingly classify one text per variable per prompt. Coding 100,000 te...

arxiv.org

It’s the final day of Week 3 at the #GESISfallseminar! Yesterday’s dinner gave us plenty of good food and lively conversations — the perfect lead-in to today. Wishing everyone success as you put your new skills into practice, and looking forward to starting the last week on Monday!

⚡Just back from the ECREA PolComm Section in beautiful Innsbruck! 📊Presentation: Trust & Influencers leading up to the 2024 U.S. election. 🌱Huge thanks to my colleagues @ikmz.bsky.social for encouraging me through my 1st conference presentation & @everyone for creating a welcoming environment!