why LSTM or RNN based methods cannot do bag-level relation extraction

[ShellingFord221]

Hi, I wonder why LSTM or RNN based methods cannot tackle bag-level relation extraction? It seems like they can only tackle sentence-level relation classification, but I think CNN or LSTM is only the encoder of sentence to get representation. There is always a way to resemble them to get bag's representation. Thanks!

[gaotianyu1350]

LSTM or RNN can definitely tackle bag-level RE. It's just that the performance gap is very small and CNN is a lot more efficient than LSTM.