Signed persistence diagrams provide a concise representation of the rank invariant (persistent Betti numbers) or birth-death invariant for filtrations of topological spaces. When these filtrations are indexed by a linear poset, the signed persistence diagrams consist of nonnegative values, allowing for straightforward interpretation as the multiplicity of the bars in the barcode of the filtration. However, for filtrations indexed by non-linear posets, signed persistence diagrams can include negative values, which complicates their interpretability. In this talk, we introduce Grassmannian persistence diagrams, a notion that extends the capabilities of signed persistence diagrams while preserving interpretability. Moreover, when the indexing poset is linear, Grassmannian persistence diagrams also reveal canonical cycle spaces for each bar in the barcode associated with the filtration.