Wikipedia Quality

All about concepts, researches and services related to quality assessment of the Multilingual @Wikipedia.

[WikiResearch @WikiResearch]

"Specificity Bias and the Wikipedia Paradox: An xAI Analysis of AI Text Detectors under Domain Shift" - purely statistical detection paradigms face systematic limitations in highly structured human domains such as Wikipedia

(Morloher et al, 2026)

https://t.co/1HO5zG7Fx5

[WikiResearch @WikiResearch]

"WPTokens: A Dataset for Studying the Regulation of Wikipedia" Sentence- and node-level change tracking over the full revision history of French Wikipedia

(@leojoubert, 2026)

https://t.co/Xxy7d2o8CB

[Włodzimierz Lewoniewski @Lewoniewski]

From #Wikipedia to Map of #SpeculativeFiction: PUEB Researchers Reach the @Pyrkon 2027 Competition Final.
https://t.co/hEdTCEi5lk

[WikiResearch @WikiResearch]

"When the Commons Appropriates a Large Language Model: How WikiVault Reshaped Korean Wikipedia" amplifying production capacity, shifting work to reviewing articles, and importing editorial judgments from enwiki
https://t.co/gKlhibJunr

@song_inhwa_, @sohwng, T.Yoo, @manoelribeiro

[KIE UEP @kie_uep]

🔍 How can AI help assess Wikipedia quality at scale?

At the conference “Open Knowledge: #Wikimedia & Research in the CEE Region”, Dr. Włodzimierz Lewoniewski presented research on #Wikipedia content, information sources, data visualizations.

More info: https://t.co/KPHGotG0V5

[WikiResearch @WikiResearch]

"Beyond Initial Creation: Two Decades of Community-based Chart Revisions from Wikimedia Commons" a novel dataset of 7,786 longitudinal chart revisions across 1,664 images.

(Shih et al, 2026)

https://t.co/Y47sehw2vL

[Włodzimierz Lewoniewski @Lewoniewski]

📏 Measuring #Wikipedia Quality at Scale: Assessment of Content, Information Sources, Data Visualizations. 👨‍🏫 Presented at #OpenKnowledge: #Wikimedia & #Research in the CEE Region.
🧩 Slides: https://t.co/UtiHIHY4xA

[WikiResearch @WikiResearch]

"WikiFlow: Predicting Wikipedia Editor Dropout at Scale Using Distributed Machine Learning on Google Cloud" https://t.co/jlp42po88z
#paywall💰
4 "editor archetypes":
Frustrated burst editors
Casual contributors ("most valuable")
Power/talk-page editors
Focused/bot-like accounts

[WikiResearch @WikiResearch]

"Linking Speakers of the German Parliament to Wikidata: Scope and Coverage of Metadata" https://t.co/ccJhALzU9x
For Bundestag members, "Wikidata is almost complete" for basic properties like gender or DOB, but coverage of "CV" fields like "educated at" is "significantly weaker"

[KIE UEP @kie_uep]

📢 Invited Talk by @Lewoniewski at @ASU:
Methods and Tools for Automatic Quality Assessment of #Wikipedia Articles and Their Information Sources.
🔗 More information: https://t.co/dQQlYVeFQI

[WikiResearch @WikiResearch]

"Can Language Models Play the Wikipedia Game?" (get from one article to another via article links) https://t.co/Yl4SqVXhIR
"No model significantly outperformed humans, although the best models (GPT-o3 mini, GPT-4o, QwQ:32B, and DeepSeek-R1:70B) achieved comparable success."

[Science&Wiki @wiki_science]

📚🌍 Open Knowledge: Wikimedia & Research in the CEE Region
📍📅 Katowice, Poland, 8–10 October 2026

👉https://t.co/7AhaKDzg2A

📸 CC BY-SA 4.0 Darafsh, Alexmar983 2026 Wikimedia Commons

#ResearchCollaboration #WikimediaResearch #WikimediaCEE
#OpenKnowledge #OpenScience

[WikiResearch @WikiResearch]

"WikiCoach: Scaffolding Learning for Novices in Online Apprenticeship Communities" https://t.co/ovhik7cYwE
"a Q&A–style LLM system designed to scaffold learning-through-doing in Wikipedia"

[KIE UEP @kie_uep]

🤖 Artificial Intelligence in #Sentiment Analysis of Multilingual #Wikipedia: #Classifiers vs. Large Language Models.
🔓 #OpenAccess research by Dr. Milena Stróżyna, Dr. Włodzimierz Lewoniewski, Izabela Czumałowska.
📖 More details: https://t.co/aai34PqDXR

[WikiResearch @WikiResearch]

"WiCleanData: Guaranteeing the Type Consistency of Wikidata by Taxonomy Refinement and Constraint Enforcement" https://t.co/Nmu0QZA8bI
Pipeline "cleans the taxonomy with [LLM] assistance, then simplifies type constraints by hierarchical aggregation, and finally filters facts"

[Włodzimierz Lewoniewski @Lewoniewski]

🧩 A new interactive map of the most popular articles on English #Wikipedia is now available. You can use this tool to search for articles, explore them as colourful puzzle pieces, zoom in, and trace how readers move from one article to another. https://t.co/BEzXTBTaqJ

[WikiResearch @WikiResearch]

"The Evolution of Coordination in a Collective Intelligence System: 25 Years of English Wikipedia and the Emergence of Generative AI" https://t.co/MN0mFuUY0G
"participation in coordination spaces has declined relative to content production, particularly in governance areas"

[BestRef @BestRefNet]

📢 #BestRef July 2026 update!
🔬4.2 million websites from more than 400 million #Wikipedia references across 300+ language editions were analyzed.
🌍 This update provides a unique view of the sources @Wikipedia relies on most.
🔍 Try it now: https://t.co/4V7DQD3nGa

[Nicholas Decker @captgouda24]

When Wikipedia is edited, real science is affected. The articles cited by it themselves get cited more, and it can even change the direction of innovation. Full paper below:

[Włodzimierz Lewoniewski @Lewoniewski]

🌐 Which websites matter most as information sources on #Wikipedia?

📊 More than 400M citations from 68M articles across 300+ language editions were analyzed, taking into account both citation frequency and article popularity.

🚀 Explore more: https://t.co/GW4FWAZyFn