How Grounded is Wikipedia? A Study on Structured Evidential Support and Retrieval

Read full text
Authors: William Gantt Walden, Kathryn Ricci, Miriam Wanner, Zhengping Jiang, Chandler May, Rong Zhou, Benjamin Van Durme
Cited by: 0

Abstract

Wikipedia is a critical resource for modern NLP, serving as a rich repository of up-to-date and citation-backed information on a wide variety of subjects.The reliability of Wikipediaits groundedness in its cited sources-is vital to this purpose.This work analyzes both how grounded Wikipedia is in this sense and how effectively fine-grained grounding evidence can be retrieved.To this end, we introduce PEOPLEPROFILES 1 -a large-scale, multi-level dataset of claim support annotations on biographical Wikipedia articles.We show that: (1) ∼22% of claims in Wikipedia lead sections are unsupported by the article body; (2) ∼30% of claims in the article body are unsupported by their (public) sources; and (3) real-world Wikipedia citation practices often differ from documented standards.Finally, we show that complex evidence retrieval remains a challenge, even for recent reasoning rerankers.

Fields of Study & Subjects

Topic Modeling Wikis in Education and Collaboration Advanced Graph Neural Networks

Publication details

Journal / source
Findings of the Association for Computational Linguistics: ACL 2026
Publisher
Association for Computational Linguistics
Pages
400–420
Publication date
1 January 2026
Language
EN