The COMPTEXT Association

How can we better capture marginalised voices through computational text analysis? An inspiring roundtable at #COMPTEXT2025 sparked important reflections. Huge thanks to @AhrabhiKat, Angelie Kraft, and @indiiigosky for guiding this vital conversation!

Thanks #COMPTEXT2025 for a great conference in #Vienna! I presented (1) a new method -- "MQ-Class" -- for LLM-classifying the full spectrum of media ideology and (2) my project mapping the 2D structure of global media ideology. Great people, panels, and most importantly food!

Even though COMPTEXT2025 closed last weekend, its spirit will live on in the splendid photography of our dedicated group of colleagues from the University of Vienna: Jakob-Moritz Eberl, @AMPlanitzer, @apeksha_sh! We thank and appreciate their work!

Had an amazing time at my first #COMPTEXT2025 in Vienna. Happy to see so much cool research, get the chance to present our paper and represent a growing team of researchers from @IPS_FSV_CUNI

After 3 packed days of research, workshops, and networking, #COMPTEXT2025 has officially wrapped up! In 48+ panels, 220+ presentations, and a full day of tutorials we received a great review of the state-of-the-art in our field. Thanks everyone and see you next year!

The success of COMPTEXT 2025 was made possible by the indefatigable and amazing work of the local organising team led by @FabienneLind + Hajo Boomgaarden @univienna @CompCommLab. The COMPTEXT Association is grateful for their work and to our sponsors (bit.ly/430VUye)!

We are happy to announce the recipients of the COMPTEXT Association travel grants. Grants were based on priorities such as ensuring global attendance and diversity. 2025 recipients are: Junaid Ahmed, Jiyoung Lee, Larissa Miller, Naama Rivlin, Ren Tao, Zeynep Zulfikar

A lively late morning at #COMPTEXT2025 brought stimulating sessions on narrative and framing, LLM performance, affective polarisation, political language in crisis, media bias, and protest in authoritarian regimes.

The #COMPTEXT2025 conference is back in full swing with Day 2. Our panels address multilingual text classification, LLM tools, democratic legitimacy, political discourse on crises, media polarization, and far-right digital narratives.

#COMPTEXT2025’s afternoon panels delivered thrilling presentations on validating computational methods, navigating populism in digital spaces, European crisis politics, global power shifts, and gender in political narratives. The thought-provoking topics sparked rich discussion.

#COMPTEXT2025 is thriving, with a massive crowd gathering—queues winding through the halls! Afternoon panels tackle computational method validation, political analysis, populism on social media, European policy responses, global power dynamics, and gender in political discourse.

I have always suspected that large language models need energy. But courtesy of long-needed research by @patrickparschan @seanpalicki and colleagues we now have more solid evidence of the impact of using them for research purposes #COMPTEXT2025I

If you have heard enough what LLM can do at #COMPTEXT2025, let us tell you what LLMs CANNOT do in our panel: We don’t need no LLMs (Sat 2:45pm BIG Hörsaal). I will present my work on the linguistic and contextual bias that I found in a boatload of LLMs (over 120 setups).

#COMPTEXT2025 is charged with insights as parallel panels explore key themes like automated text extraction, multimodal political communication, climate politics, domestic economic framing, tech ethics, and social identities.

The biggest ever #COMPTEXT conference at the University of Vienna is officially under way! Registration has opened and the first panels—on ideological estimation, misinformation, climate discourse, global economic narratives, and digital content influence—are now taking place.

Afternoon workshops at #COMPTEXT2025 wrapped up with in-depth sessions on narrative analysis, platform access (DSA), concept measurement, and LLM training. Thanks to instructors Maria Antoniak, @dscheykopp, @hauke_licht,@chklamm and all participants for a great end to the day!

A stimulating morning of hands-on tutorials at #COMPTEXT2025 introduced participants to advanced techniques in digital trace data collection, visual content analysis, and Bayesian text modelling. The day is far from over — stay tuned for more cutting-edge methods in action.

The registration room is buzzing and the first courses have already started! #COMPTEXT 2025 is officially under way at the University of Vienna!

Rooms are packed to capacity as tutorials on cutting-edge AI research techniques unfold at #COMPTEXT2025 Valerie Hase / @FriederRodewald / Tobias Heidenreich @hdnrch / Petro Tolochko

Looking forward to #COMPTEXT2025 in Vienna this weekend! I’ll be presenting a paper on a new NLP+LLM-based approach to measuring rule design—who is regulated, how, and on what terms—applied to EU environmental directives and transposition delays. comptextconference.org/7th-annual-com…