Juri Opitz

Thanks a lot for the explanations! Seems like it's impossible to address well. Also thanks for the proposed solutions, I will try both things. Should I close this issue? Am...

> Maybe in a future iteration we can develop a uber alignment suite for amrlib that includes multiple and malleable functionality. That's a good suggestion, thanks for your thoughts.

I also think that it has something to do with the pre-processing from https://github.com/sheng-z/stog. When predicting the test sentences from LDC everything seems to work fine, it's just for new...

Thanks @mkartik But that doesn't explain why it would work well for the **test partition of LDC**, does it? Anyways, may I ask you if you have tried out this...

Thanks a lot @mkartik for this helpful investigation! Yes, there is a significant overlap (unfortunately) between train NEs and test NEs. But I think it still does not explain everything,...

Yes, that's what I mean. I am very sorry for the complicated example. I simplified the example a lot now, I hope it is now more clear. Again, I'm sorry...

Yes it is very interesting, and I agree that the bug is not exactly in your parser but in the pre-processing in [stog](https://github.com/sheng-z/stog). There are also some similar issues reported...

@jcyk I can confirm that it works so much better with the non GR model. Thanks for your quick advice and again, I congratulate you for your very awesome work!...

@Zoher15 What you posted is not a bug in penman and I also see no reason why it shouldn't be a valid AMR (it's just redundancy). So this is an...

The relevant material had been linked by @kfnks in this [comment](https://github.com/acl-org/acl-anthology/issues/2791#issuecomment-1926848848). To quote: https://www.dropbox.com/scl/fi/wu74zqzgdp34p0ifreznd/papers.yml?rlkey=uqtttxtbnuzrl4tl1jlwfqjs1&dl=0