How Grounded is Wikipedia? A Study on Structured Evidential Support and Retrieval
This work addresses the problem of verifying information reliability in Wikipedia for NLP researchers and users, but it is incremental as it builds on existing concerns about citation practices.
The study analyzed Wikipedia's reliability by measuring how well claims are supported by citations, finding that about 22% of claims in lead sections lack support from the article body and 30% from publicly accessible sources, and that evidence retrieval remains difficult for current AI models.
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 Wikipedia -- its groundedness in its cited sources -- is vital to this purpose. This work analyzes both how grounded Wikipedia is and how readily fine-grained grounding evidence can be retrieved. To this end, we introduce PeopleProfiles -- 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 publicly accessible 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.