Questions about aggregation method

[RapperYao]

Hello, thank you so much for providing such excellent work and code. However, as I was reading the SPIN paper, I noticed that you did a weighted aggregation of messages within one neighbor at different time steps, but in the last part, you did not do a weighted aggregation of messages from all neighbors. Just as the influence of the messages is different at different time steps, the influence of different neighbors should be different, why didn't you take this into account? I would appreciate it very much if you could answer it.