Figure 2
From: Multipartite state generation in quantum networks with optimal scaling

The figure depicts the different schemes we consider. (a) Starting with multiple copies of a multipartite state (GHZ states in this example) each station implements the multipartite hashing protocol (marked by the orange box) in a measurement-based way to obtain fewer copies of that state with higher fidelity. (b) Before the protocol neighboring stations share several copies of Bell pairs. The stations implement the bipartite hashing protocol (orange boxes) in a measurement-based way to obtain purified Bell pairs. Then, in a separate step, the stations prepare another resource state for merging the Bell-pairs (blue box) into the desired multipartite state. (c) The neighboring stations share, like in scheme B, several copies of noisy Bell pairs. Each station performs the measurement based implementation of the multipartite hashing protocol (orange boxes) or the hashing protocol combined with the merging operation (purple box) where connections are needed.