On the Hodge-GUE correspondence