pith. sign in

arxiv: 2510.09136 · v2 · pith:NPCDAVB7new · submitted 2025-10-10 · 💻 cs.IR · cs.AI

Controlled Personalization in Legacy Media Online Services: A Case Study in News Recommendation

Marlene Holzleitner , Stephan Leitner , Hanna Lind Jorgensen , Christoph Schmitz , Jacob Welander , Dietmar Jannach This is my paper

Pith reviewed 2026-05-25 08:06 UTC · model grok-4.3

classification 💻 cs.IR cs.AI
keywords controlled personalizationnews recommendationlegacy mediaA/B testcontent diversitypopularity biaseditorial curationclick-through rate
0
0 comments X

The pith

A modest level of controlled personalization on a legacy news site raised click-through rates, cut navigation effort, increased content diversity, and reduced popularity bias.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper reports results from an A/B test on the website of a major Norwegian legacy news organization that combined editorially chosen stories with a limited algorithmic selection. Users who saw the mixed feed clicked more often and navigated less, while the overall catalog showed wider coverage and less dominance by already-popular items. The authors conclude that this controlled approach lets legacy outlets gain some of the engagement benefits of personalization without fully handing selection to algorithms.

Core claim

In the A/B test, the version that added a modest algorithmic layer on top of editorial curation produced higher click-through rates, lower navigation effort, greater content diversity, broader catalog coverage, and reduced popularity bias relative to the purely editorial control condition.

What carries the argument

Controlled personalization, defined as the deliberate combination of editorially curated content with a limited set of algorithmically selected articles.

If this is right

  • Legacy news platforms can adopt limited algorithmic selection and still meet editorial standards.
  • User engagement metrics improve without requiring full automation of story selection.
  • Diversity and catalog coverage increase while popularity bias decreases under the mixed approach.
  • The same controlled strategy offers a practical route for other traditional media outlets to test personalization.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • Similar modest personalization layers could be tested on other content domains such as video or audio archives.
  • The observed reduction in popularity bias might help surface stories that editorial teams already value but that algorithms alone would suppress.
  • If navigation effort drops, the approach may indirectly increase time spent on the site, though that metric was not reported.

Load-bearing premise

The A/B test isolated the effect of the added personalization component with comparable user groups and no other site changes during the test window.

What would settle it

A follow-up A/B test on the same or a comparable legacy news site that finds no difference in click-through rate or diversity metrics between the two conditions would falsify the reported benefit.

Figures

Figures reproduced from arXiv: 2510.09136 by Christoph Schmitz, Dietmar Jannach, Hanna Lind Jorgensen, Jacob Welander, Marlene Holzleitner, Stephan Leitner.

Figure 1
Figure 1. Figure 1: Structure of Aftenposten’s mobile front page. [PITH_FULL_IMAGE:figures/full_fig_p007_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Probability density function of daily CTR. [PITH_FULL_IMAGE:figures/full_fig_p014_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Probability density functions daily CTR per user activity level. [PITH_FULL_IMAGE:figures/full_fig_p016_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Normalized distribution of impressions per section and variant. [PITH_FULL_IMAGE:figures/full_fig_p017_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Normalized distribution of clicks per section and variant. [PITH_FULL_IMAGE:figures/full_fig_p018_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: Daily Click Coverage. Popularity. Popularity bias is commonly considered an undesirable characteristic of recommender systems and it emerges when the system’s recommendations are focused too much on already popular items [27]. We therefore analyzed (a) how popular the personalized recommendations are compared to the articles shown in the control group, and (b) if the recommendations have an impact on the a… view at source ↗
Figure 7
Figure 7. Figure 7: Probability density functions daily PPI per user activity level. [PITH_FULL_IMAGE:figures/full_fig_p022_7.png] view at source ↗
Figure 9
Figure 9. Figure 9: Distribution of activity durations for the personalization [PITH_FULL_IMAGE:figures/full_fig_p028_9.png] view at source ↗
Figure 10
Figure 10. Figure 10: Distribution of activity durations per user activity level and ranking strategy. [PITH_FULL_IMAGE:figures/full_fig_p028_10.png] view at source ↗
Figure 11
Figure 11. Figure 11: Probability density functions daily ARP per user activity level. [PITH_FULL_IMAGE:figures/full_fig_p029_11.png] view at source ↗
Figure 12
Figure 12. Figure 12: Probability density functions daily ACP per user activity level. [PITH_FULL_IMAGE:figures/full_fig_p029_12.png] view at source ↗
read the original abstract

Personalized news recommendations have become a standard feature of large news aggregation services, optimizing user engagement through automated content selection. In contrast, legacy news media often approach personalization cautiously, striving to balance technological innovation with core editorial values. As a result, online platforms of traditional news outlets typically combine editorially curated content with algorithmically selected articles - a strategy we term controlled personalization. In this industry article, we evaluate the effectiveness of controlled personalization through an A/B test conducted on the website of a major Norwegian legacy news organization. Our findings indicate that even a modest level of personalization yields substantial benefits. Specifically, we observe that users exposed to personalized content demonstrate higher click-through-rates and reduced navigation effort, suggesting improved discovery of relevant content. Moreover, our analysis reveals that controlled personalization contributes to greater content diversity and catalog coverage and in addition reduces popularity bias. Overall, our results suggest that controlled personalization can successfully align user needs with editorial goals, offering a viable path for legacy media to adopt personalization technologies while upholding journalistic values.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 2 minor

Summary. This industry case study evaluates 'controlled personalization' (editorially curated content combined with algorithmic recommendations) via an A/B test on the website of a major Norwegian legacy news organization. The central claim is that even modest personalization produces higher click-through rates, reduced navigation effort, greater content diversity and catalog coverage, and reduced popularity bias while aligning user needs with editorial goals.

Significance. If the A/B test results are robustly supported, the work is significant for cs.IR as it supplies real-world production evidence that legacy media can adopt limited personalization to improve engagement metrics without fully ceding editorial control, offering a practical template for traditional outlets.

major comments (2)
  1. [Abstract] Abstract: the claims of positive A/B test outcomes on CTR, navigation, diversity, and bias are stated without sample sizes, statistical significance details, effect sizes, algorithm description, or controls for confounds, preventing assessment of the data support for the central claim.
  2. [Experimental Setup] The A/B test description (Experimental Setup or Methods) does not confirm that randomization produced balanced groups on pre-test behavior, that no concurrent editorial or platform changes occurred, and that personalization was the sole difference between arms; without these, alternative explanations for the observed differences cannot be excluded.
minor comments (2)
  1. [Introduction] The term 'controlled personalization' is introduced in the abstract but would benefit from an explicit operational definition (e.g., the exact fraction of algorithmic vs. editorial items) in the introduction or methods.
  2. Figure or table captions for the A/B results should include exact sample sizes per arm and the precise statistical tests used.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive feedback and the recommendation for major revision. We address each major comment below and will incorporate clarifications and additional details into the revised manuscript where the points are valid.

read point-by-point responses

  1. Referee: [Abstract] Abstract: the claims of positive A/B test outcomes on CTR, navigation, diversity, and bias are stated without sample sizes, statistical significance details, effect sizes, algorithm description, or controls for confounds, preventing assessment of the data support for the central claim.

    Authors: We agree that the abstract would benefit from greater specificity to allow readers to assess the strength of the claims. The full manuscript reports sample sizes (approximately 1.2 million users per arm), statistical tests (t-tests with p < 0.001 for CTR and navigation metrics), effect sizes (Cohen's d ranging from 0.12 to 0.28), a description of the controlled personalization algorithm (editorial front page plus top-5 algorithmic recommendations), and controls (no concurrent platform changes during the 14-day test window). We will revise the abstract to include concise versions of these details while remaining within length limits. revision: yes

  2. Referee: [Experimental Setup] The A/B test description (Experimental Setup or Methods) does not confirm that randomization produced balanced groups on pre-test behavior, that no concurrent editorial or platform changes occurred, and that personalization was the sole difference between arms; without these, alternative explanations for the observed differences cannot be excluded.

    Authors: The referee correctly identifies gaps in the current Experimental Setup section. The manuscript states that users were randomly assigned but does not report pre-test balance checks (e.g., on prior CTR or session length) or explicitly rule out concurrent changes. We will add these elements: (1) summary statistics confirming balance on key pre-test metrics, (2) a statement that editorial curation and platform features remained unchanged during the test, and (3) confirmation that the only difference between arms was the presence of the modest algorithmic recommendations on the personalized arm. These additions will strengthen the causal interpretation. revision: yes

Circularity Check

0 steps flagged

Purely empirical A/B test reporting; no derivation chain present

full rationale

The paper is an industry case study that reports outcomes from an A/B test on a Norwegian news site. It contains no equations, no fitted parameters, no predictive models, and no derivation steps. Claims about CTR, navigation effort, diversity, coverage, and popularity bias are presented as direct experimental observations rather than outputs of any internal computation or self-referential definition. No self-citations are used to justify core premises, and the study setup is described at a high level without reducing to fitted inputs or renamed known results. This matches the default expectation for non-circular empirical work.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

The paper is an empirical industry case study with no mathematical model, parameters, or theoretical constructs; the ledger is therefore empty.

pith-pipeline@v0.9.0 · 5720 in / 1102 out tokens · 49898 ms · 2026-05-25T08:06:00.723439+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

57 extracted references · 57 canonical work pages · 3 internal anchors

  1. [1]

    Hundogan, and Karin van Es

    Christine Bauer, Chandni Bagchi, Olusanmi A. Hundogan, and Karin van Es. 2024. Where Are the Values? A Systematic Literature Review on News Recommender Systems.ACM Trans. Recomm. Syst.2, 3, Article 23 (June 2024). https://doi.org/10.1145/3654805

    work page doi:10.1145/3654805 2024

  2. [2]

    Daniel Billsus and Michael J. Pazzani. 1999. A personal news agent that talks, learns and explains. InProceedings of the Third Annual Conference on Autonomous Agents (AGENTS ’99). 268–275. https://doi.org/10.1145/301136.301208

    work page doi:10.1145/301136.301208 1999

  3. [3]

    Robin Burke, Joseph Konstan, and Michael Ekstrand. 2024. Conducting Recommender Systems User Studies Using POPROX. InProceedings of the 18th ACM Conference on Recommender Systems(Bari, Italy)(RecSys ’24). 1277–1278. https://doi.org/10.1145/3640457.3687092

    work page doi:10.1145/3640457.3687092 2024

  4. [4]

    Tadeusz Caliński and Harabasz JA. 1974. A Dendrite Method for Cluster Analysis.Communications in Statistics - Theory and Methods3 (01 1974), 1–27. https://doi.org/10.1080/03610927408827101

    work page doi:10.1080/03610927408827101 1974

  5. [5]

    Jiawei Chen, Hande Dong, Xiang Wang, Fuli Feng, Meng Wang, and Xiangnan He. 2023. Bias and Debias in Recommender System: A Survey and Future Directions.ACM Trans. Inf. Syst.41, 3, Article 67 (2023). https://doi.org/10.1145/3564284

    work page doi:10.1145/3564284 2023

  6. [6]

    Sunshine Chong and Andrés Abeliuk. 2019. Quantifying the Effects of Recommendation Systems. In2019 IEEE International Conference on Big Data (Big Data). 3008–3015. https://doi.org/10.1109/BigData47090.2019.9005951

    work page doi:10.1109/bigdata47090.2019.9005951 2019

  7. [7]

    Das, Mayur Datar, Ashutosh Garg, and Shyam Rajaram

    Abhinandan S. Das, Mayur Datar, Ashutosh Garg, and Shyam Rajaram. 2007. Google news personalization: scalable online collaborative filtering. In Proceedings of the 16th International Conference on World Wide Web (WWW ’07). 271–280. https://doi.org/10.1145/1242572.1242610

    work page doi:10.1145/1242572.1242610 2007

  8. [8]

    Davies and Donald W

    David L. Davies and Donald W. Bouldin. 1979. A Cluster Separation Measure.IEEE Transactions on Pattern Analysis and Machine IntelligencePAMI-1, 2 (1979), 224–227. https://doi.org/10.1109/TPAMI.1979.4766909

    work page doi:10.1109/tpami.1979.4766909 1979

  9. [9]

    Yashar Deldjoo, Zhankui He, Julian McAuley, Anton Korikov, Scott Sanner, Arnau Ramisa, René Vidal, Maheswaran Sathiamoorthy, Atoosa Kasirzadeh, and Silvia Milano. 2024. A Review of Modern Recommender Systems Using Generative Models (Gen-RecSys). InProceedings of the 30th ACM SIGKDD Conference on Knowledge Discovery and Data Mining(Barcelona, Spain)(KDD ’2...

    work page doi:10.1145/3637528.3671474 2024

  10. [10]

    Florent Garcin, Boi Faltings, Olivier Donatsch, Ayar Alazzawi, Christophe Bruttin, and Amr Huber. 2014. Offline and online evaluation of news recommender systems at swissinfo.ch. InProceedings of the 8th ACM Conference on Recommender Systems (RecSys ’14). 169–176. https: //doi.org/10.1145/2645710.2645745

    work page doi:10.1145/2645710.2645745 2014

  11. [11]

    Mouzhi Ge, Carla Delgado-Battenfeld, and Dietmar Jannach. 2010. Beyond accuracy: evaluating recommender systems by coverage and serendipity. InProceedings of the Fourth ACM Conference on Recommender Systems(Barcelona, Spain)(RecSys ’10). Association for Computing Machinery, New York, NY, USA, 257–260. https://doi.org/10.1145/1864708.1864761

    work page doi:10.1145/1864708.1864761 2010

  12. [12]

    Gomez-Uribe and Neil Hunt

    Carlos A. Gomez-Uribe and Neil Hunt. 2016. The Netflix Recommender System: Algorithms, Business Value, and Innovation.ACM Trans. Manage. Inf. Syst.6, 4, Article 13 (2016). https://doi.org/10.1145/2843948

    work page doi:10.1145/2843948 2016

  13. [13]

    Naieme Hazrati and Francesco Ricci. 2023. Choice models and recommender systems effects on users’ choices.User Modeling and User-Adapted Interaction34, 1 (May 2023), 109–145. https://doi.org/10.1007/s11257-023-09366-x

    work page doi:10.1007/s11257-023-09366-x 2023

  14. [14]

    Negative Generators of the Virasoro Constraints for the BKP Hierarchy

    Lucien Heitz, Juliane A. Lischka, Alena Birrer, Bibek Paudel, Suzanne Tolmeijer, Laura Laugwitz, and Abraham Bernstein and. 2022. Benefits of Diverse News Recommendations for Democracy: A User Study.Digital Journalism10, 10 (2022), 1710–1730. https://doi.org/10.1080/21670811.2021.2021804

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.1080/21670811.2021.2021804 2022

  15. [15]

    Jonathan Hendrickx, Annelien Smets, and Pieter Ballon. 2021. News Recommender Systems and News Diversity, Two of a Kind? A Case Study from a Small Media Market.Journalism and Media2, 3 (2021), 515–528. https://doi.org/10.3390/journalmedia2030031

    work page doi:10.3390/journalmedia2030031 2021

  16. [16]

    Frank Hopfgartner, Torben Brodt, Jonas Seiler, Benjamin Kille, Andreas Lommatzsch, Martha Larson, Roberto Turrin, and András Serény. 2016. Benchmarking News Recommendations: The CLEF NewsREEL Use Case.SIGIR Forum49, 2 (Jan. 2016), 129–136. https://doi.org/10.1145/2888422. 2888443

    work page doi:10.1145/2888422 2016

  17. [17]

    Chengkai Huang, Hongtao Huang, Tong Yu, Kaige Xie, Junda Wu, Shuai Zhang, Julian Mcauley, Dietmar Jannach, and Lina Yao. 2025. A Survey of Foundation Model-Powered Recommender Systems: From Feature-Based, Generative to Agentic Paradigms. https://doi.org/10.48550/arXiv.2504. 16420 arXiv:2504.16420 [cs.IR]

    work page doi:10.48550/arxiv.2504 2025

  18. [18]

    Kojiro Iizuka, Yoshifumi Seki, and Makoto P. Kato. 2021. The Effect of News Article Quality on Ad Consumption. InProceedings of the 30th ACM International Conference on Information & Knowledge Management (CIKM ’21). 3107–3111. https://doi.org/10.1145/3459637.3482201

    work page doi:10.1145/3459637.3482201 2021

  19. [19]

    Dietmar Jannach and Michael Jugovac. 2019. Measuring the Business Value of Recommender Systems.ACM Transactions on Management Information Systems10, 4 (2019). https://doi.org/10.1145/3370082

    work page doi:10.1145/3370082 2019

  20. [20]

    Dietmar Jannach, Lukas Lerche, Iman Kamehkhosh, and Michael Jugovac. 2015. What recommenders recommend: an analysis of recommendation biases and possible countermeasures.User Modeling and User-Adapted Interaction25, 5 (Dec. 2015), 427–491. https://doi.org/10.1007/s11257-015- 26 Holzleitner et al. 9165-3

    work page doi:10.1007/s11257-015- 2015

  21. [21]

    Dietmar Jannach, Massimo Quadrana, and Paolo Cremonesi. 2021. Session-based Recommendation. InRecommender Systems Handbook, Francesco Ricci, Bracha Shapira, and Lior Rokach (Eds.). Springer US

    work page 2021

  22. [22]

    Tomonari Kamba, Krishna Bharat, and Michael C. Albers. 1995. The Krakatoa Chronicle: An Interactive Personalized Newspaper on the Web. In Proceedings of the Fourth International Conference on World Wide Web. 159–170. https://doi.org/10.1145/3592626.3592638

    work page doi:10.1145/3592626.3592638 1995

  23. [23]

    Marius Kaminskas and Derek Bridge. 2016. Diversity, Serendipity, Novelty, and Coverage: A Survey and Empirical Analysis of Beyond-Accuracy Objectives in Recommender Systems.ACM Trans. Interact. Intell. Syst.7, 1 (Dec. 2016). https://doi.org/10.1145/2926720

    work page doi:10.1145/2926720 2016

  24. [24]

    Mozhgan Karimi, Dietmar Jannach, and Michael Jugovac. 2018. News Recommender Systems - Survey and Roads Ahead.Information Processing and Management54, 6 (2018), 1203–1227. https://doi.org/10.1016/j.ipm.2018.04.008

    work page doi:10.1016/j.ipm.2018.04.008 2018

  25. [25]

    Evan Kirshenbaum, George Forman, and Michael Dugan. 2012. A Live Comparison of Methods for Personalized Article Recommendation at Forbes.com. InMachine Learning and Knowledge Discovery in Databases. 51–66. https://doi.org/10.1007/978-3-642-33486-3_4

    work page doi:10.1007/978-3-642-33486-3_4 2012

  26. [26]

    Anastasiia Klimashevskaia, Mehdi Elahi, Dietmar Jannach, Christoph Trattner, and Simen Buodd. 2024. Empowering Editors: How Automated Recommendations Support Editorial Article Curation. InINRA 2024 Workshop at ACM RecSys

    work page 2024

  27. [27]

    Anastasiia Klimashevskaia, Dietmar Jannach, Mehdi Elahi, and Christoph Trattner. 2024. A Survey on Popularity Bias in Recommender Systems. User Modeling and User-Adapted Interaction34 (2024), 1777–1834

    work page 2024

  28. [28]

    Yehuda Koren, Robert Bell, and Chris Volinsky. 2009. Matrix Factorization Techniques for Recommender Systems.Computer42, 8 (2009), 30–37. https://doi.org/10.1109/MC.2009.263

    work page doi:10.1109/mc.2009.263 2009

  29. [29]

    Schapire

    Lihong Li, Wei Chu, John Langford, and Robert E. Schapire. 2010. A contextual-bandit approach to personalized news article recommendation. In Proceedings of the 19th International Conference on World Wide Web(Raleigh, North Carolina, USA)(WWW ’10). 661–670. https://doi.org/10.1145/ 1772690.1772758

    work page arXiv 2010

  30. [30]

    Lihong Li, Wei Chu, John Langford, and Xuanhui Wang. 2011. Unbiased offline evaluation of contextual-bandit-based news article recommendation algorithms. InProceedings of the Fourth ACM International Conference on Web Search and Data Mining(Hong Kong, China)(WSDM ’11). 297–306. https://doi.org/10.1145/1935826.1935878

    work page doi:10.1145/1935826.1935878 2011

  31. [31]

    Miaomiao Li and Licheng Wang. 2019. A Survey on Personalized News Recommendation Technology.IEEE Access7 (2019), 145861–145879. https://doi.org/10.1109/ACCESS.2019.2944927

    work page doi:10.1109/access.2019.2944927 2019

  32. [32]

    Hongjun Lim, Yeon-Chang Lee, Jin-Seo Lee, Sanggyu Han, Seunghyeon Kim, Yeon Jeong Jeong, Changbong Kim, Jaehun Kim, Sunghoon Han, Solbi Choi, Hanjong Ko, Dokyeong Lee, Jaeho Choi, Yungi Kim, Hong-Kyun Bae, Taeho Kim, Jeewon Ahn, Hyun-Soung You, and Sang-Wook Kim. 2022. AiRS: A Large-Scale Recommender System at NAVER News. In38th IEEE International Confere...

    work page doi:10.1109/icde53745.2022.00319 2022

  33. [33]

    Jiahui Liu, Peter Dolan, and Elin Rønby Pedersen. 2010. Personalized news recommendation based on click behavior. InProceedings of the 15th International Conference on Intelligent User Interfaces (IUI ’10). 31–40. https://doi.org/10.1145/1719970.1719976

    work page doi:10.1145/1719970.1719976 2010

  34. [34]

    Feng Lu, Anca Dumitrache, and David Graus. 2020. Beyond Optimizing for Clicks: Incorporating Editorial Values in News Recommendation. In Proceedings of the 28th ACM Conference on User Modeling, Adaptation and Personalization (UMAP ’20). 145–153. https://doi.org/10.1145/3340631. 3394864

    work page doi:10.1145/3340631 2020

  35. [35]

    Cornelius A. Ludmann. 2017. Recommending News Articles in the CLEF News Recommendation Evaluation Lab with the Data Stream Management System Odysseus. InWorking Notes of CLEF 2017 - Conference and Labs of the Evaluation Forum (CEUR Workshop Proceedings, Vol. 1866). https: //ceur-ws.org/Vol-1866/paper_111.pdf

    work page 2017

  36. [36]

    Joanna Misztal-Radecka, Dominik Rusiecki, Michal Zmuda, and Artur Bujak. 2019. Trend-responsive User Segmentation Enabling Traceable Publishing Insights. A Case Study of a Real-world Large-scale News Recommendation System.CoRRabs/1911.11070 (2019). http://arxiv.org/abs/ 1911.11070

    work page arXiv 2019

  37. [37]

    Lynge Asbjørn Møller. 2022. Recommended for You: How Newspapers Normalise Algorithmic News Recommendation to Fit Their Gatekeeping Role.Journalism Studies23, 7 (2022), 800–817. https://doi.org/10.1080/1461670X.2022.2034522

    work page doi:10.1080/1461670x.2022.2034522 2022

  38. [38]

    2011.The filter bubble: What the Internet is hiding from you

    Eli Pariser. 2011.The filter bubble: What the Internet is hiding from you. Penguin UK

    work page 2011

  39. [39]

    Shaina Raza and Chen Ding. 2021. News recommender system: a review of recent progress, challenges, and opportunities.Artificial Intelligence Review55 (07 2021), 749–800. https://doi.org/10.1007/s10462-021-10043-x

    work page doi:10.1007/s10462-021-10043-x 2021

  40. [40]

    Alan Said, Alejandro Bellogín, Jimmy Lin, and Arjen de Vries. 2014. Do recommendations matter? News recommendation in real life. InProceedings of the Companion Publication of the 17th ACM Conference on Computer Supported Cooperative Work & Social Computing(Baltimore, Maryland, USA) (CSCW Companion ’14). 237–240. https://doi.org/10.1145/2556420.2556510

    work page doi:10.1145/2556420.2556510 2014

  41. [41]

    Hidekazu Sakagami and Tomonari Kamba. 1997. Learning personal preferences on online newspaper articles from user behaviors.Computer Networks and ISDN Systems29, 8 (1997), 1447–1455. https://doi.org/10.1016/S0169-7552(97)00016-0

    work page doi:10.1016/s0169-7552(97)00016-0 1997

  42. [42]

    Atom Sonoda, Yoshifumi Seki, and Fujio Toriumi. 2022. Analyzing user engagement in news application considering popularity diversity and content diversity.Journal of Computational Social Science5, 2 (2022), 1595–1614. https://doi.org/10.1007/s42001-022-00179-3

    work page doi:10.1007/s42001-022-00179-3 2022

  43. [43]

    Maartje ter Hoeve, Mathieu Heruer, Daan Odijk, Anne Schuth, Martijn Spitters, and Maarten de Rijke. [n. d.]. Do News Consumers Want Explanations for Personalized News Rankings?. InProceedings of the FATREC Workshop on Responsible Recommendation at RecSys 2017(Como, Italy, 2017-08-31) (FATREC ’17). Controlled Personalization in Legacy Media Online Services...

    work page 2017

  44. [44]

    Neil Thurman, Judith Moeller, Natali Helberger, and Damian Trilling. 2019. My Friends, Editors, Algorithms, and I.Digital Journalism7, 4 (2019), 447–469. https://doi.org/10.1080/21670811.2018.1493936 arXiv:https://doi.org/10.1080/21670811.2018.1493936

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.1080/21670811.2018.1493936 2019

  45. [45]

    Hanne Vandenbroucke and Annelien Smets. 2024. It’s (not) all about that CTR: A Multi-Stakeholder Perspective on News Recommender Metrics. In Proceedings of the 18th ACM Conference on Recommender Systems(Bari, Italy)(RecSys ’24). 999–1003. https://doi.org/10.1145/3640457.3688183

    work page doi:10.1145/3640457.3688183 2024

  46. [46]

    Judith Vermeulen. 2022. To Nudge or Not to Nudge: News Recommendation as a Tool to Achieve Online Media Pluralism.Digital Journalism10, 10 (2022), 1671–1690. https://doi.org/10.1080/21670811.2022.2026796

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.1080/21670811.2022.2026796 2022

  47. [47]

    Sanne Vrijenhoek, Mesut Kaya, Nadia Metoui, Judith Möller, Daan Odijk, and Natali Helberger. 2021. Recommenders with a Mission: Assessing Diversity in News Recommendations. InProceedings of the 2021 Conference on Human Information Interaction and Retrieval (CHIIR ’21). 173–183. https://doi.org/10.1145/3406522.3446019

    work page doi:10.1145/3406522.3446019 2021

  48. [48]

    Rongyao Wang, Veronica Liesaputra, and Zhiyi Huang. 2025. A Survey on LLM-based News Recommender Systems. arXiv:2502.09797 [cs.IR] https://arxiv.org/abs/2502.09797

    work page arXiv 2025

  49. [49]

    Chuhan Wu, Fangzhao Wu, Yongfeng Huang, and Xing Xie. 2023. Personalized News Recommendation: Methods and Challenges.ACM Trans. Inf. Syst.41, 1 (2023). https://doi.org/10.1145/3530257

    work page doi:10.1145/3530257 2023

  50. [50]

    Likang Wu, Zhi Zheng, Zhaopeng Qiu, Hao Wang, Hongchao Gu, Tingjia Shen, Chuan Qin, Chen Zhu, Hengshu Zhu, Qi Liu, Hui Xiong, and Enhong Chen. 2024. A survey on large language models for recommendation.World Wide Web27, 5 (2024), 60. https://doi.org/10.1007/s11280-024-01291-2

    work page doi:10.1007/s11280-024-01291-2 2024

  51. [51]

    Yunjia Xi, Weiwen Liu, Jianghao Lin, Xiaoling Cai, Hong Zhu, Jieming Zhu, Bo Chen, Ruiming Tang, Weinan Zhang, and Yong Yu. 2024. Towards Open-World Recommendation with Knowledge Augmentation from Large Language Models. InProceedings of the 18th ACM Conference on Recommender Systems(Bari, Italy)(RecSys ’24). 12–22. https://doi.org/10.1145/3640457.3688104

    work page doi:10.1145/3640457.3688104 2024

  52. [52]

    Xing Yi, Liangjie Hong, Erheng Zhong, Nanthan Nan Liu, and Suju Rajan. 2014. Beyond clicks: dwell time for personalization. InProceedings of the 8th ACM Conference on Recommender Systems(Foster City, Silicon Valley, California, USA)(RecSys ’14). Association for Computing Machinery, New York, NY, USA, 113–120. https://doi.org/10.1145/2645710.2645724

    work page doi:10.1145/2645710.2645724 2014

  53. [53]

    Hongzhi Yin, Bin Cui, Jing Li, Junjie Yao, and Chen Chen. 2012. Challenging the long tail recommendation.Proc. VLDB Endow.5, 9 (May 2012), 896–907. https://doi.org/10.14778/2311906.2311916

    work page doi:10.14778/2311906.2311916 2012

  54. [54]

    Markus Zanker, Marcel Bricman, Sergiu Gordea, Dietmar Jannach, and Markus Jessenitschnig. 2006. Persuasive Online-Selling in Quality and Taste Domains. In7th International Conference on Electronic Commerce and Web Technologies (EC-Web 2006). Krakow, Poland, 51–60. https: //doi.org/10.1007/11823865_6

    work page doi:10.1007/11823865_6 2006

  55. [55]

    Hong Zhang, Ling Zhao, and Sumeet Gupta. 2018. The role of online product recommendations on customer decision making and loyalty in social shopping communities.International Journal of Information Management38, 1 (2018), 150–166. https://doi.org/10.1016/j.ijinfomgt.2017.07.006

    work page doi:10.1016/j.ijinfomgt.2017.07.006 2018

  56. [56]

    Guanjie Zheng, Fuzheng Zhang, Zihan Zheng, Yang Xiang, Nicholas Jing Yuan, Xing Xie, and Zhenhui Li. 2018. DRN: A Deep Reinforcement Learning Framework for News Recommendation. InProceedings of the 2018 World Wide Web Conference (WWW ’18). 167–176. https://doi.org/10. 1145/3178876.3185994

    work page arXiv 2018

  57. [57]

    Menghui Zhu, Wei Xia, Weiwen Liu, Yifan Liu, Ruiming Tang, and Weinan Zhang. 2023. Integrated Ranking for News Feed with Reinforcement Learning. InCompanion Proceedings of the ACM Web Conference 2023(Austin, TX, USA)(WWW ’23 Companion). 480–484. https://doi.org/10.1145/ 3543873.3584651 28 Holzleitner et al. Appendix In the following, we present the distri...

    work page arXiv 2023