Digital knowledge platforms increasingly shape how cultural heritage and historical responsibility are represented and remembered, yet they are vulnerable to subtle forms of narrative manipulation that rarely appear as explicit misinformation. We present a new dataset of English Wikipedia revision pairs designed to support the systematic study of cultural heritage manipulation in the context of the Russo--Ukrainian conflict. The resource captures fine-grained textual changes through which cultural attribution, agency, and legitimacy are incrementally reframed within a collaborative encyclopedic infrastructure. By providing paired revisions, editorial context, and consistent annotations, the dataset enables analysis of discursive interventions that remain procedurally legitimate while producing cumulative cultural effects. The dataset is released to support research in information retrieval, natural language processing, and computational social science on detecting and understanding subtle narrative change in large-scale, collaborative knowledge systems.