๐จ Our work on machine learning for facial emotion detection in political performances is now available on arXiv!
We processed 77 hours of video and found that populist leaders express negative emotions more often than their non-populist counterparts.
https://t.co/mGI6WavQ61
The AI-GENIE package for automatic item generation and structural validation using LLMs and Exploratory Graph Analysis methods is now updated on GitHub!
Link below!
@enriquemoreno@the_qml_guy@thsottiaux Useful while we wait. Remote connections feature introduced 24h ago works out of the box. The only problematic feature is probably computer use, didn't test that.
@enriquemoreno@the_qml_guy@thsottiaux this is working very well for me: https://t.co/aStVDYJgDH
has automatic updates, even remote ssh conncections are working well
@fcoury if i accidentally quit a session (kill terminal for example) with /goal running, it cannot be restored with /resume not visible in session list.
can be restored only with session id
Applications are open for the Petnica Summer Institute Summer School on Scientific Machine Learning!
Petnica is only a few kilometers from my hometown, so I am especially happy to see this program there. A great opportunity to learn about ML in an amazing setting.
If you are an MSc student, early-stage PhD candidate, or advanced undergraduate interested in machine learning and scientific research, I would strongly encourage you to apply.
Applications are open for the Petnica Summer Institute Summer School on Scientific Machine Learning!
Petnica is only a few kilometers from my hometown, so I am especially happy to see this program there. A great opportunity to learn about ML in an amazing setting.
If you are an MSc student, early-stage PhD candidate, or advanced undergraduate interested in machine learning and scientific research, I would strongly encourage you to apply.
Grateful to the @OSFramework for pushing the SCORE initiative and publishing three important Nature papers on reproducibility, robustness, and replicability in the social and behavioural sciences.
Glad to have contributed to this paper as an analyst. In the robustness study, 100 papers were independently re-analysed, and only about one third reproduced the original result within a narrow tolerance, although most reanalyses still recovered the original qualitative conclusion, with substantial variation in effect sizes across plausible analytic choices.
A wake-up call for quantitative methods in the social sciences. Multiverse analysis will likely become increasingly important for uncertainty quantification and for making researcher degrees of freedom more explicit.
https://t.co/s5xXu8I2XM
The Ultimate Tutorial for AI-driven Scale Development in Generative Psychometrics: Releasing AIGENIE from its Bottle, is now available on @arxiv_org !! Learn how to use our AIGENIE R Package for Automatic Item Generation and Validation with Network Integrated Evaluation! Check it out! Another gem led by @larafromUVA !
Very interesting new paper by Mariana Teles and Aleksandar Tomaลกeviฤ (@atomasevic), recently published at College Teaching (link below). They used our transforEmotion R package to apply LLM-based text classification via zero-shot classification of open-ended student feedback, revealing large effect sizes across engagement and learning dimensions THAT TRADITIONAL LIKERT SCALES OF STUDENTS EVALUATION OF TEACHING missed! Very cool and innovative paper: https://t.co/Ekd7P8D7Uf
I'm thrilled to announce the launch of our Generative Psychometrics website, a new frontier where large language models meet psychometric science. Link Below!
Our interdisciplinary team treats language as a scalable source of psychological information: generating, evaluating, and structuring measurement instruments through the convergence of AI and network psychometrics.
We're addressing AI issues with Psychometrics, and Psychometric issues with AI - laying the groundwork for the Psychology of Artificial Minds.
Explore our research, open-source tools (like the `AI-GENIE` and `transforEmotion` R packages), and upcoming workshops at ๐ https://t.co/hkc06rg95n
#Psychometrics #ArtificialIntelligence #NetworkScience #OpenSource #ComputationalPsychology
Interested in identifying and modeling emotions in videos, images, and text data via AI models?Check out our new transforEmotion: An Open-Source R Package for Emotion Analysis Using Transformer-Based Generative AI Models
Aleksandar Tomaลกeviฤ (@atomasevic), Hudson Golino, & Alexander Christensen
Computational Communication Research, Volume 8, Issue 2, Jan 2026, p. 1
DOI: https://t.co/wg7sL43Td2
Great job by @claudeai on the new visual explanations feature.
If you ask it to explain the voter model, it even spawns a small interactive agent-based model.