V. Yadav); 1613-0073 mendation and Analytics (INRA'24) Benjamin Kille benjamin.u.kille@ntnu.no 0 1 2 Andreas Lommatzsch andreas.lommatzsch@dai-labor.de 0 1 3 Célina Treuiller 0 1 4 Vandana Yadav vandana.yadav@ntnu.no 0 1 2 Özlem Özgobek 0 1 2 INRA'24: International Workshop on News Recommendation and Analytics Mahmood , Mehdi Elahi, Farhad Vadiee, Samia Touileb and Lubos Steskal Norwegian University of Science and Technology , Sem Saerlands Vei 5 7034 Trondheim , Norway Technische Universität Berlin , Ernst-Reuter-Platz 7, 10587 Berlin , Germany University of Lorraine , 34 Cours Léopold 54000 Nancy , France 2009 000 0 0003

Personalization has changed how we engage with news. The INRA workshop provides a forum to researchers, practitioners, and interested parties to discuss recent trends concerning news personalization. This edition of INRA highlights a variety of topics including generative AI, fake news, and multi-modality. Motivation and Background Recommender systems are an essential part of today's digital news ecosystem. Nevertheless, the task of recommending news has not been solved satisfactorily. While the technological aspects have been addressed, many research questions concerning the increasing automation, efects on societies, and user experience remain unanswered. How can users continue trusting in media in a time when AI allows everyone to create texts instantly? How will users in the future engage with news? Will they switch to more multi-media channels or focus on texts? How can publishers assure a diverse news diet including important information that is hard to digest? How will social media afect the spread of news? Answering many of these questions has to involve actors from diferent research disciplines.

 news news personalization paper formatting -

Italy. CEUR ceur-ws.org

 2. Accepted Contributions

The call for participation motivated researchers to submit their findings in the form of regular or short papers. Fourteen groups of authors submitted their work. A thorough peer-review found eight of the fourteen manuscript suited for presentation: 1. A Supervised Machine Learning Approach for Supporting Editorial Article Selection, by Bilal 2. Non-Stationary Multi-Armed Bandits for News Recommendations, by Noah Daniëls and Bart

Goethals.

 3. Program Committee

We recruited a set of experts to assure the quality of the accepted papers. We would like to thank:

 3. Peeling Back the Layers: An In-Depth Evaluation of Encoder Architectures in Neural News Recommenders, by Andreea Iana, Goran Glavaš and Heiko Paulheim . 4. Negativity Sells ? Using an LLM to Afectively Reframe News Articles in a Recommender System, by Jia Hua Jeng , Gloria Kasangu, Alain D. Starke , Erik Knudsen and Christoph Trattner . 5. Empowering Editors: How Automated Recommendations Support Editorial Article Curation , by Anastasiia Klimashevskaia, Mehdi Elahi, Dietmar Jannach, Christoph Trattner and Simen Buodd . 6. RADio-: a Simplified Codebase for Evaluating Normative Diversity in Recommender Systems , by Sanne Vrijenhoek. 7. Simulating Real-World News Consumption: Deep Q-Learning for Diverse User-Centric Slate Recommendations, by Aayush Roy, Elias Tragos, Aonghus Lawlor and Neil Hurley . 8. Enhancing Prediction Models with Reinforcement Learning, by Karol Radziszewski and Piotr Ociepka .